CN117556643B

CN117556643B - Flood early warning and forecasting method and forecasting system

Info

Publication number: CN117556643B
Application number: CN202410045258.7A
Authority: CN
Inventors: 王兆吉; 高婕; 赵国辉; 冯同飞; 高晨; 郑灿
Original assignee: Hebei Baoding Hydrological Survey And Research Center
Current assignee: Hebei Baoding Hydrological Survey And Research Center
Priority date: 2024-01-12
Filing date: 2024-01-12
Publication date: 2024-03-19
Anticipated expiration: 2044-01-12
Also published as: CN117556643A

Abstract

The invention relates to the technical field of flood prediction informatization, in particular to a flood early warning and forecasting method and a forecasting system. The method comprises the following steps: acquiring water rain condition data and rainfall forecast data; hydrodynamic processing and water flow extraction are carried out on the water rain condition data and the rainfall forecast data, so that hydrodynamic data and water flow data are respectively obtained; obtaining city building distribution data, and carrying out city infiltration calculation on the city building distribution data to obtain city infiltration rate data, wherein the city infiltration rate data comprises natural infiltration rate data and artificial infiltration rate data; constructing an urban surface hydrological model according to urban infiltration rate data, hydrodynamic data and water flow data; and generating surface water area condition data according to the urban surface hydrologic model so as to perform flood early warning operation. According to the method, a more accurate urban surface hydrologic model is built, the accuracy of flood prediction is improved, and early warning is more reliable.

Description

Flood early warning and forecasting method and forecasting system

Technical Field

The invention relates to the technical field of flood prediction informatization, in particular to a flood early warning and forecasting method and a forecasting system.

Background

The flood early warning and forecasting method is a systematic method for monitoring, predicting and informing the occurrence of flood in advance so as to take proper measures to reduce the damage of flood to human beings, property and environment. Flood forecasting refers to application science and technology for revealing and predicting occurrence of flood and change process of flood according to information such as hydrology, weather and the like in the early stage and the current stage. The method is one of important contents of flood control non-engineering measures, and is directly used for flood control emergency, reasonable utilization and protection of water resources, hydraulic engineering construction and scheduling application management and industrial and agricultural safety production service. With the flood early warning system, emergency institutions can be prepared in advance, rescue work is coordinated, flood incidents can be responded more effectively, and confusion and delay are reduced. In the current flood early warning and forecasting method, the accuracy of flood prediction is insufficient due to weather mutation, inaccurate model parameters and data loss.

Disclosure of Invention

The invention provides a flood early warning and forecasting method and a forecasting system for solving at least one technical problem.

The application provides a flood early warning and forecasting method, which comprises the following steps:

Step S1: acquiring water rain condition data and rainfall forecast data;

step S2: hydrodynamic processing and water flow extraction are carried out on the water and rain condition data and the rainfall forecast data to respectively obtain hydrodynamic data and water flow data, wherein the hydrodynamic data comprise river channel hydrodynamic data and urban hydrodynamic data, the urban hydrodynamic data comprise surface hydrodynamic data and underground hydrodynamic data, and the water flow data comprise urban water flow data and river channel water flow data;

step S3: obtaining city building distribution data, and carrying out city infiltration calculation on the city building distribution data to obtain city infiltration rate data, wherein the city infiltration rate data comprises natural infiltration rate data and artificial infiltration rate data;

step S4: constructing an urban surface hydrological model according to urban infiltration rate data, hydrodynamic data and water flow data;

step S5: and generating surface water area condition data according to the urban surface hydrologic model so as to perform flood early warning operation.

According to the method, through the water rain condition data, the rainfall forecast data, the hydrodynamic force data, the water flow data and the urban building distribution data, a more accurate urban surface hydrologic model can be built, the accuracy of flood prediction is improved, and early warning is more reliable. By separating urban hydrodynamic data and river hydrodynamic data, the method can monitor hydrodynamic conditions in the city, including surface hydrodynamic and underground hydrodynamic, in real time, so that urban managers can better cope with potential flood risks. By acquiring city building distribution data and calculating city infiltration rate, the method considers the influence of city infiltration, and more accurately simulates the hydrologic process in the city, thereby improving the accuracy of flood prediction. By integrating the hydrologic models of the city and the river, the interaction between the city and the river is considered, so that flood prediction is more comprehensive, and the mechanism of urban flood formation is better understood. By generating surface water area condition data, the method can monitor flood conditions in real time, trigger a flood early warning system when necessary, and take emergency measures early so as to reduce damage caused by flood. The method provides city hydrologic data and models, and can be used for city planning, water resource management and emergency response.

Preferably, step S1 is specifically:

step S11: acquiring original water and rain condition data through a database connected with a hydrologic station to obtain the original water and rain condition data;

step S12: acquiring original rainfall forecast data through a database connected with a weather bureau to obtain the original rainfall forecast data;

step S13: and carrying out data preprocessing on the original water and rain condition data and the original rainfall forecast data to obtain the water and rain condition data and the rainfall forecast data.

According to the method, through connecting the hydrologic station with the database of the meteorological bureau, the original water rain condition data and rainfall forecast data of various data sources can be obtained, and the diversity and the comprehensiveness of the data are ensured. The data base is connected, so that real-time data acquisition of hydrologic stations and weather stations can be realized, the latest water and rain condition data and rainfall forecast data can be ensured to be acquired, and the timely response to weather changes is facilitated. The data preprocessing comprises the steps of missing value filling, abnormal value detection, correction and the like, and is beneficial to reducing data errors and improving the reliability and accuracy of data. The original data are converted into the formats of the water rain condition data and the rainfall forecast data, so that the data are easier to use by a system and a model, and the usability and operability of the data are improved.

Preferably, step S13 is specifically:

step S131: carrying out data cleaning on the original water and rain condition data and the original rainfall forecast data to obtain original water and rain condition cleaning data and original rainfall forecast cleaning data;

step S132: carrying out integrity evaluation on the original water rain condition cleaning data and the original rainfall forecast cleaning data to obtain integrity evaluation data, wherein the integrity evaluation data comprise missing evaluation data and non-missing evaluation data, and the missing evaluation data comprise space missing evaluation data and time missing evaluation data;

step S133: when the integrity evaluation data are determined to be the missing evaluation data, interpolation processing is carried out on the original water rain condition cleaning data and the original rainfall forecast cleaning data according to the missing evaluation data to obtain the water rain condition data and the rainfall forecast data;

wherein the step of interpolation processing includes the steps of:

step S134: when the missing evaluation data are determined to be the space missing evaluation data, performing spatial interpolation processing on the original water and rain condition cleaning data and the original rainfall forecast cleaning data to respectively obtain the water and rain condition data and the rainfall forecast data;

step S135: when determining that the missing evaluation data is time missing evaluation data, performing time interpolation processing on the original water and rain condition cleaning data and the original rainfall forecast cleaning data to obtain water and rain condition data and rainfall forecast data;

Step S136: when the missing evaluation data are determined to be space-time missing evaluation data, performing space-time interpolation processing on the original water and rain condition cleaning data and the original rainfall forecast cleaning data to obtain water and rain condition data and rainfall forecast data respectively.

According to the invention, through data cleaning (step S131), noise, abnormal values and inconsistencies in the original data can be removed, and the quality and the credibility of the data are improved, so that errors introduced in subsequent analysis are reduced. Through the integrity assessment (step S132), the missing values in the data can be determined, which are divided into spatial missing and temporal missing, helping to identify the cause and scope of the data missing. When the data missing property evaluation determines that the data is missing, the interpolation process (step S133) may select an appropriate interpolation method according to the missing property of the data. And interpolating the space deficiency evaluation data, such as the space deficiency in the remote sensing data, so as to obtain the water rain condition data and the rainfall forecast data of the deficiency area. The time-missing evaluation data, such as time-missing due to instrument failure, is interpolated to fill in missing values in the time series. And carrying out space-time interpolation on space-time missing evaluation data, such as space-time missing in a remote sensing time sequence, so as to fill in missing water and rain condition data and rainfall forecast data. The continuity of the data can be maintained through interpolation processing, so that the data sequence is kept smooth in time and space, and the model establishment and analysis are facilitated. The interpolated water and rain condition data and rainfall forecast data have more complete time and space information, so that the usability and operability of the data are enhanced, and more accurate flood early warning and forecasting are facilitated.

Preferably, step S134 is specifically:

carrying out data division on the original water and rain condition cleaning data according to the water and rain condition place position data corresponding to the original water and rain condition cleaning data to obtain water and rain condition place division data, and carrying out data division on the original rainfall forecast cleaning data according to the rainfall place position data corresponding to the original rainfall forecast cleaning data to obtain rainfall place division data;

carrying out data density calculation on the water and rain condition site division data and the rainfall site division data to respectively obtain water and rain condition site data density data and rainfall site data density data, wherein the water and rain condition site data density data comprises high water and rain condition site data density data and low water and rain condition site data density data, and the rainfall site data density data comprises high rainfall site data density data and low rainfall site data density data;

when the water and rain condition place data density data/rainfall place data density data are determined to be high water and rain condition place data density data/high rainfall place data density data, performing inverse distance weight processing on original water and rain condition cleaning data/original rainfall forecast cleaning data to obtain water and rain condition data/rainfall forecast data;

When the water and rain condition place data density data/rainfall place data density data are determined to be low water and rain condition place data density data/low rainfall place data density data, performing uncertainty radial basis function interpolation processing on original water and rain condition cleaning data/original rainfall forecast cleaning data to obtain water and rain condition data/rainfall forecast data;

wherein the step of uncertainty radial basis function interpolation processing comprises the steps of:

performing random input interpolation processing on the original water rain condition cleaning data/the original rainfall forecast cleaning data according to a preset radial basis function to obtain first random input interpolation data;

performing random input interpolation processing on the original water rain condition cleaning data/the original rainfall forecast cleaning data according to a preset semi-variation function to obtain second random input interpolation data;

acquiring historical water and rain condition data/historical rainfall data, and performing confidence assessment on the first random input interpolation data and the second random input interpolation data according to the historical water and rain condition data/the historical rainfall data to obtain confidence assessment data;

generating water rain condition data/rainfall forecast data according to the confidence evaluation data, the first random input interpolation data and the second random input interpolation data.

According to the invention, the original water rain condition cleaning data and the rainfall forecast cleaning data are divided into different places, and the data density is calculated, so that the high-density data and the low-density data can be distinguished, a proper interpolation method can be selected more finely, and the adaptability of the data is improved. According to the division of the data density, different interpolation methods, such as inverse distance weight interpolation and uncertainty radial basis function interpolation, can be flexibly selected, and the characteristics of processing data of different places better are facilitated. For high-density data, the inverse distance weight interpolation is adopted, so that more accurate data estimation can be provided, and the data quality and accuracy are improved, especially in a high-density data area. For low-density data, uncertainty radial basis function interpolation is adopted, the uncertainty of the data is considered, the spatial variation and irregular distribution of the data can be processed better, and more reliable data estimation is provided. By using the historical water rain condition data/the historical rainfall data to carry out confidence evaluation on the interpolation result, the reliability of the interpolation result can be quantized, the reliability of the interpolation result can be understood by a user, and the reliability of data use can be improved. The interpolation method is helpful to maintain the time-space continuity of the data, so that the generated water rain condition data/rainfall forecast data is smoothly transited in time and space, and is easier to use by a model and a system.

Preferably, step S135 specifically includes:

when the missing evaluation data are determined to be time missing evaluation data, linear interpolation processing is carried out on the original water rain condition cleaning data and the original rainfall forecast cleaning data to obtain first time interpolation data;

performing space distribution Kriging interpolation processing on the original water rain condition cleaning data and the original rainfall forecast cleaning data to obtain second time interpolation data;

carrying out weighted fusion on the first time interpolation data and the second time interpolation data to obtain water and rain condition data and rainfall forecast data;

wherein the step of spatially distributed kriging interpolation processing comprises the steps of:

performing spatial variation processing on the original water rain condition cleaning data and the original rainfall forecast cleaning data to obtain spatial variation data;

carrying out local spatial trend identification according to the spatial variation data to obtain local spatial trend data;

and performing Kriging interpolation processing on the local spatial trend data, the original water rain condition cleaning data and the original rainfall forecast cleaning data to obtain second time interpolation data.

According to the invention, through linear interpolation (the first time interpolation data in the step S135), missing values in a time sequence can be filled, so that the water and rain condition data and rainfall forecast data are ensured to be continuous in time, and the data integrity is maintained. The deficiency value can be estimated according to the spatial variation of the data by using the spatial distribution kriging interpolation (the second time interpolation data in step S135), and the kriging interpolation method considers the spatial relationship, so that the spatial distribution characteristics of the data can be better reflected. And carrying out weighted fusion on the first time interpolation data and the second time interpolation data, and generating more accurate and reliable water and rain condition data and rainfall forecast data by utilizing the results of different time interpolation. By adopting time interpolation and space distribution Kriging interpolation, continuity of the generated water rain condition data and rainfall forecast data can be ensured, and the model establishment and analysis are facilitated. The interpolation method fills up the missing value in the data, improves the availability of the data, and enables the location and the time point of the missing data to be used for building and analyzing the flood early warning model. Because the interpolation method provides more complete and fine water rain condition data and rainfall forecast data, the accuracy of the flood early warning system is improved, the flood risk is identified in advance, and necessary measures are taken.

Preferably, step S2 is specifically:

carrying out unit hydrologic unit depth processing on the water and rain condition data and the rainfall forecast data to obtain water flow data;

finite element division is carried out on the water flow data to obtain water flow division data;

carrying out boundary condition processing on the water flow dividing data to obtain water flow grid data;

and carrying out hydrodynamic calculation on the water flow grid data to obtain hydrodynamic data.

According to the invention, through unit hydrologic unit advanced treatment (the first step in the step S2), the water rain condition data and the rainfall forecast data are converted into water flow data, so that the rainfall information is converted into input data required by a flood model, and the data are more operable. Finite element division of the water flow data (second step in step S2) can divide the complex water flow system into smaller units for numerical simulation and calculation to more accurately simulate the real water flow situation. The boundary condition processing (the third step in the step S2) is performed on the water flow dividing data, so that the influence of the boundary condition on the water flow is considered by the model, and the fidelity of the model is improved, particularly when urban hydrodynamic force is simulated. Through hydrodynamic calculation (fourth step in step S2), the behavior of the water flow, including water level, flow rate, flow direction, etc., can be simulated, helping to understand the dynamic process of the water flow. Due to finite element partitioning and hydrodynamic calculations, the generated hydrodynamic data is typically of high resolution, which can provide detailed water flow conditions, helping to predict floods more accurately. The processing and calculation in the step S2 are beneficial to improving the precision and accuracy of the flood model, so that the flood early warning system has higher reliability, and the possibility of false alarm and missing alarm is reduced.

Preferably, step S3 is specifically:

obtaining city building distribution data, wherein the city building distribution data comprises artificial building material data and natural soil type data;

calculating the material infiltration rate according to the artificial building material data to obtain artificial infiltration rate data;

and (3) carrying out natural infiltration rate calculation according to the natural soil type data to obtain natural infiltration rate data.

The method for acquiring the urban building distribution data (comprising the artificial building material data and the natural soil type data) is a key step for establishing the urban infiltration rate model, and the properties and the constitution of urban ground surfaces and the influence of urbanization on the infiltration rate are more accurately considered. And (3) calculating the infiltration rate of the materials according to the artificial building material data (the second step in the step S3), so that the model is helpful to know the infiltration properties of different building materials in the city. And (3) carrying out natural infiltration rate calculation according to the natural soil type data (the third step in the step S3), which is helpful for the model to consider the infiltration characteristics of the natural soil and accurately simulate the hydrologic process of the urban ground surface and the natural ground surface. By combining the artificial infiltration rate data and the natural infiltration rate data, an urban infiltration rate model can be constructed, and the model considers the characteristics of urban ground surfaces, such as roads, buildings and the like, and soil types, so that the infiltration capacity of the ground surfaces can be estimated better. The urban infiltration rate is one of important parameters of flood simulation, and the accuracy of flood simulation can be improved and the flood condition can be predicted better by considering the characteristics of urban earth surfaces and the infiltration rates of different soil types.

Preferably, step S4 is specifically:

acquiring pipeline drainage data;

marking data and carrying out terrain classification according to the pipeline drainage data and the urban building distribution data to obtain urban terrain classification data;

performing space-time simulation according to the water flow data and the urban terrain division data to obtain hydrological space-time simulation data;

performing infiltration treatment on urban infiltration rate data and hydrologic space-time simulation data to obtain hydrologic space-time infiltration data;

and carrying out runoff treatment on the hydrodynamic data and the hydrologic space-time infiltration data to obtain the urban surface hydrologic model.

The acquisition of the pipeline drainage data is one of key data for constructing the urban surface hydrologic model, and the data is helpful for knowing the performance and the capacity of the urban drainage system, so that the flood condition of the city can be better simulated. The marking of the data and the division of the terrain are carried out according to the pipeline displacement data and the urban building distribution data (the second step in the step S4), so that the urban ground surface is divided into different areas, and the urban terrain characteristics, the position of the drainage system and the displacement are considered. By performing space-time simulation (the third step in step S4) using the water flow data and the urban topography division data, the space-time distribution of the water flow in the city can be simulated, which is helpful for understanding the spreading and evolution process of the flood. And (3) performing infiltration treatment on the urban infiltration rate data and the hydrological space-time simulation data (the fourth step in the step S4), so that the infiltration capacity of the urban ground surface is taken into consideration, the influence of rainfall on the surface runoff is reduced, and the surface hydrological process is simulated more accurately. The dynamic behavior of urban surface water flow can be simulated by carrying out runoff treatment (the fifth step in the step S4) through hydrodynamic data and hydrological space-time infiltration data, and the method is very important for flood simulation and flood prediction. The processing and calculating method in the step S4 is finally used for constructing an urban surface hydrologic model for flood prediction and analysis, and is beneficial to improving flood management and emergency response capability of the city.

Preferably, step S5 is specifically:

generating surface water area conditions according to the urban surface water model to obtain surface water area condition data, wherein the surface water area condition data comprise surface water area peak value change data and surface water area depth change data;

and carrying out flood grade generation and flood range generation on the surface water area condition data to obtain flood grade data and flood range data so as to carry out flood early warning operation.

According to the method, the surface water condition data (comprising the surface water peak value change data and the surface water depth change data) are generated according to the urban surface water model, so that the water condition of the urban surface can be known in real time, and important information about the change condition of flood on the urban surface is provided. By analyzing and processing the surface water area condition data, flood grade data can be generated, which is helpful for determining the severity of flood, so that the risk and influence of flood can be better evaluated. According to flood grade data and surface water area condition data, flood range data can be generated, the determination of the area affected by flood is facilitated, and the operation of a flood early warning system is supported. The generated flood grade data and flood range data can be used for flood early warning operation, and once flood risks or critical conditions are detected, the system can trigger early warning and take corresponding emergency measures to reduce the influence of flood on cities and residents. The result of the step S5 is helpful for the urban flood management department to better understand and respond to flood events, improves the flood management capacity of the city, and is helpful for reducing the loss caused by flood.

Preferably, the present application further provides a flood early-warning and forecasting system, configured to execute the flood early-warning and forecasting method as described above, where the flood early-warning and forecasting system includes:

the hydrologic data acquisition module is used for acquiring water rain condition data and rainfall forecast data;

the system comprises a hydrodynamic flow acquisition module, a water flow acquisition module and a control module, wherein the hydrodynamic flow acquisition module is used for carrying out hydrodynamic processing and water flow extraction on water rain condition data and rainfall forecast data to obtain hydrodynamic data and water flow data, the hydrodynamic data comprise river channel hydrodynamic data and urban hydrodynamic data, the urban hydrodynamic data comprise surface hydrodynamic data and underground hydrodynamic data, and the water flow data comprise urban water flow data and river channel water flow data;

the urban infiltration calculating module is used for acquiring urban building distribution data, and carrying out urban infiltration calculation on the urban building distribution data to obtain urban infiltration rate data, wherein the urban infiltration rate data comprises natural infiltration rate data and artificial infiltration rate data;

the urban surface hydrological model construction module is used for constructing an urban surface hydrological model according to urban infiltration rate data, hydrodynamic data and water flow data;

And the flood early warning module is used for generating surface water area condition data according to the urban surface hydrologic model so as to perform flood early warning operation.

The invention has the beneficial effects that: the invention covers the acquisition, processing and extraction of water rain condition data, rainfall forecast data, hydrodynamic force data and water flow data, comprehensively analyzes a large amount of water and weather data, and provides reliable basic data for flood forecast. And S4, constructing a city surface hydrological model by utilizing city infiltration rate data, hydrodynamic force data and water flow data, wherein the model considers factors such as city topography, underground hydrodynamic force, infiltration rate and the like, and can more accurately simulate the hydrological process in the city. Step S5 generates surface water volume condition data, including surface water volume peak change data and surface water volume depth change data, through the urban surface water model, providing detailed information about the distribution and change of floods on the urban surface. The invention can more comprehensively understand the formation and development process of flood by comprehensively processing multi-source data including meteorological data, hydrological data and urban building data, and is helpful for improving flood monitoring and understanding hydrological conditions. The generated surface water area condition data and the urban surface water model can be used for flood early warning operation, and once flood risks are detected, the system can trigger alarms and take emergency measures, such as evacuation, embankment reinforcement and the like, so as to reduce the influence of flood on cities and residents.

Drawings

Other features, objects and advantages of the present application will become more apparent upon reading of the detailed description of non-limiting implementations made with reference to the following drawings in which:

FIG. 1 is a flow chart showing the steps of a flood warning and forecasting method according to an embodiment;

FIG. 2 shows a flow chart of the steps of a hydrological data collection method of an embodiment;

FIG. 3 is a flow chart showing the steps of a method for preprocessing hydrologic raw data of an embodiment;

fig. 4 shows a flow chart of the steps of an interpolation processing method of an embodiment.

Detailed Description

The following is a clear and complete description of the technical method of the present patent in conjunction with the accompanying drawings, and it is evident that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, are intended to fall within the scope of the present invention.

Furthermore, the drawings are merely schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. The functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor methods and/or microcontroller methods.

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

Referring to fig. 1 to 4, the present application provides a flood early warning and forecasting method, which includes the following steps:

step S1: acquiring water rain condition data and rainfall forecast data;

specifically, real-time rainfall data, including rainfall and rainfall intensity, is obtained by connecting weather sensors of a weather bureau.

Specifically, satellite remote sensing technology is utilized to obtain wide-area rainfall data, including rainfall cloud pictures and predicted rainfall distribution.

Specifically, historical water level, flow and rain amount data is collected from the surface hydrologic site to build a hydrologic database.

Specifically, rainfall data is input into the model using a numerical hydrologic model, such as HEC-HMS, simulating surface runoff and water flow dynamics.

Specifically, a city water flow network is deduced by using a Geographic Information System (GIS) technology and combining with the topographic data, and water flow is calculated.

Specifically, groundwater level data is acquired by using a remote sensing technology, and groundwater power is calculated by combining a groundwater model.

specifically, building profiles and land utilization data are obtained from a city planning department to determine the building type and coverage of a city.

Specifically, using the remote sensing image and the topographic data, the surface type of the city, such as roads, greenbelts, buildings, etc., is analyzed.

Specifically, urban soil type and infiltration rate data are obtained through field measurement and soil sampling. Soil types in different areas of the city are identified. This can be done by in-situ observation, soil sampling and laboratory testing, with soil types typically including sandy soil, loam, clay, and the like. For each soil type, a permeability test of the soil is performed to determine its permeability coefficient (or preset in a database based on empirical data), which represents the rate of moisture permeation per unit area per unit time, measured by using a penetrometer or hydraulic test, etc. The calculation of the natural infiltration rate will depend on the soil type and the permeability. In general, the natural hypotonic rate can be calculated by the following formula: natural downflow = permeability coefficient x rainfall intensity, where rainfall intensity is the rate of rainfall, typically expressed in terms of rainfall per unit time. The calculation of the artificial infiltration rate generally involves subtracting the natural infiltration rate from the reduction caused by the building and the artificial facilities. Expressed by the following formula: artificial infiltration rate = natural infiltration rate-artificial interference, wherein the artificial interference includes building coverage, watertight pavement, water storage facilities, and the like.

specifically, a two-dimensional hydrologic model is established based on urban surface features and hydrodynamic data, taking into account the effects of surface runoffs, groundwater flows and buildings. As first, hydrologic basis data including city map, building distribution data, terrain data, infiltration rate data, hydrodynamic data, rainfall data, etc. are collected. According to the city map and the data, the space range of the model is determined, and the city is divided into different subareas, usually based on drainage basins. Parameters are set for the urban surface hydrologic model, including surface roughness, infiltration rate, soil type, building type, road characteristics, etc., which will affect the simulation results of the water flow, infiltration and confluence processes. And establishing a surface and subsurface seepage layer in the model. The surface layer represents urban surfaces, including roads, buildings, greenbelts, etc., the subsurface layers represent subsurface soil and rock, with different infiltration rates, and the buildings and roads can be modeled as independent elements within an area, which can affect rainfall runoff and infiltration. Boundary conditions are set according to the inlet and outlet points of water flow, including the inlet and outlet of a rain water discharge, river, lake or other body of water. The historical or real-time rainfall data is used to simulate rainfall events, and the rainfall data can be divided into different time steps, and each time step has different rainfall intensity and distribution. The model is run to simulate hydrologic processes including stormwater runoff, hypotonic, groundwater flow and sewage discharge, and will take into account hydrodynamic properties of the earth's surface and the hypotonic layer, hypotonic rate, topography and the influence of the building. The model will generate hydrologic data during rainfall events, including flood depth, runoff flow, infiltration volume, etc.

Specifically, a distributed hydrological model, such as SWMM, is used to simulate urban hydrological processes including stormwater runoff, downhill, sewage discharge, etc.

Specifically, a soil hydrologic model and urban building data are combined to build the urban hydrologic model, and different infiltration rates and hydrodynamic conditions are considered.

Specifically, the urban surface water flow process is simulated according to a hydrologic model, and surface water volume condition data including water level, water depth and flow velocity distribution are generated.

Specifically, the remote sensing data and the measured water level data are combined with the hydrological model result to generate surface water area condition data.

Specifically, a geographic information system is used for analyzing the water flow path and the water level change to generate a surface water area situation map.

Specifically, the hydrologic model of the city predicts a flood event. The predictions show that during the next 24 hours, the rainfall will continue to increase, resulting in a rise in river water level. According to the model result, the warning water level is set to be 2 meters, and once the water level reaches or exceeds the level, flood warning is issued. When the water level rises to 1.8 meters, the flood warning mechanism issues warning information informing residents of flood risks and suggesting that they remain alert. When the water level reaches 2 meters, an alarm escalates, the relevant authorities/departments initiate emergency evacuation plans and require residents to evacuate the potentially disaster area. Flood prediction also shows roads, bridges and low lying areas affected by the flood, warning people to avoid these areas.

Preferably, step S1 is specifically:

specifically, the real-time data source connected to the hydrologic station acquires real-time water level, flow, rainfall and other water and rain condition data, for example, the real-time data of each hydrologic station is acquired by connecting to a database of the national hydrologic bureau.

specifically, a weather data source connected to a weather station acquires real-time weather data including information such as rainfall forecast, air temperature, humidity, etc., for example, satellite cloud image data connected to the weather station and output data of a rainfall model.

Specifically, the original water rain condition data is subjected to data cleaning and correction to remove abnormal values and error data. And carrying out time synchronization and spatial interpolation on the original rainfall forecast data so as to ensure the consistency and accuracy of the data. Generating water rain condition data and rainfall forecast data of a time sequence for subsequent processing.

Preferably, step S13 is specifically:

specifically, using the data cleaning tool, outliers, duplicates, and inconsistencies in the water rain condition and rainfall forecast data are deleted. For water rain data, data points for water level measurement errors or faults are processed. For rainfall forecast data, unreasonable predictions are processed.

Specifically, outlier processing: water rain condition data: for water rain condition data, firstly, abnormal values such as unreasonable water level or rainfall values need to be detected and processed, and one method is to use a statistical method, for example, outlier detection based on a mean value and a standard deviation, and data points which deviate from a normal range obviously are marked as abnormal values, and the abnormal values can be corrected, replaced or deleted according to the reasons of the abnormal values and the data quality standard. Rainfall forecast data: for rainfall forecast data, outliers include unreasonable extreme rainfall values, similar outlier detection methods may be used to mark outliers that are outside of reasonable ranges as outliers when processing the rainfall forecast data, including correcting outliers to maximum or minimum reasonable values, or comparing with other relevant data to determine its accuracy. Repeating the value processing: there are duplicate records in the data that need to be deleted or combined to ensure the uniqueness of the data, in water rain condition data, duplicate values are due to duplicate measurements or data collection, in rainfall forecast data, there are forecast data at the same time point of multiple forecast model generation, and these data need to be combined or screened. Inconsistent value processing: the inconsistent values in the data relate to problems in terms of units, data formats or measurement standards, and when the water and rain condition data are processed, the same units are adopted for all the data, for example, the unified water level data is meter or centimeter, for rainfall forecast data, the consistent time stamp format is ensured, and the forecast time is matched with the actual time. Missing value processing: there are missing values in the data, and methods such as interpolation are needed to process these missing values. In the water rain condition data, if the water level data at a certain time point is missing, an interpolation method can be considered to fill in the missing value, such as linear interpolation or time sequence interpolation, and in the rainfall forecast data, the forecast value of the adjacent time point can be used for estimating the value of the missing time point, such as a linear estimation mode or a smoothing processing mode.

specifically, the integrity assessment is performed, including checking for missing data, generating missing assessment data based on the time stamp and geographic location of the data, e.g., if a certain hydrologic station has no data for a certain period of time, marking it as a time missing, and if data points in a certain region are sparse, marking it as a space missing.

wherein the step of interpolation processing includes the steps of:

Specifically, spatial interpolation methods (such as kriging interpolation, inverse distance weight interpolation and the like) are used for filling in the spatial missing evaluation data, and interpolation is performed on each missing hydrologic site based on the data of surrounding sites to generate water and rain condition data.

specifically, the time missing evaluation data is filled up by using a time interpolation method (such as linear interpolation, cubic spline interpolation and the like), and the missing data of each time period is interpolated based on the data of the adjacent time period to generate water and rain condition data.

Specifically, a time-space interpolation method (such as time-space kriging interpolation) is used for filling the time-space deficiency evaluation data, and for data points with space and time deficiency, the time-space interpolation method is used for comprehensively considering data of surrounding sites and time periods to generate water and rain condition data.

Preferably, step S134 is specifically:

specifically, the original water and rain condition cleaning data are divided according to the position information of the water and rain condition places, and data sets of different places are generated. And dividing the original rainfall forecast and cleaning data according to the position information of the rainfall places.

specifically, high water rain condition site data density data: the number of data points surrounding each water rain situation site is counted to determine whether it belongs to a high density site. Data density of low water rain condition places: the number of data points surrounding each water rain site is counted to determine whether it belongs to a low density site. High rainfall site data density data: the number of data points surrounding each rainfall site is counted to determine whether it belongs to a high density site. Low rainfall site data density data: the number of data points surrounding each rainfall site is counted to determine whether it belongs to a low density site.

specifically, for high-water rain condition place data density data/high-rainfall place data density data, an inverse distance weight interpolation method is used for filling up missing values, and the method interpolates data according to the distance between data points, wherein the data points with the distance are larger in weight, and the data points with the distance are smaller in weight.

Specifically, for example, for each missing target point, the distance between it and all known data points is calculated. The distance is calculated using a selected distance metric method, the closer the distance the more weight it should be to the known data point. Distance weight calculation formula:，/>is->Weights of the neighboring points, +.>To observe the point to the->Distance of the adjacent points, +.>The weight index is positive. For each missing target point, a weighted average is performed using the values of the known data points and the calculated weights, interpolating the calculation formula: / >，/>Is at a position +.>Estimated value of->Is the observation of the i-th neighbor,/->Is the corresponding weight.

specifically, for low-water rain condition place data density data/low-rainfall place data density data, an uncertainty radial basis function interpolation method is used for filling in the missing value.

specifically, radial basis functions:，/>is a Gaussian radial basis function, +.>Is a positive constant, controls the shape of the gaussian function,/>Is the distance between the input point and the base point, when +.>Near 0, +.>Near 1, whenIn case of enlargement, the wearer is strapped with>Rapidly trending towards 0 enables a gaussian radial basis function to mathematically describe the relationship between local features and data points.

specifically, the half-variation function:，/>is a half-variation function representing the half-variation value between two points, < >>Is a noise variance or base value, which indicates the degree of variation between two points without deviation,is the distance (or spatial separation) between two points,>is the space autocorrelation distance or range, controlThe decay rate of the half-variant function. The exponential half-variation function describes that the correlation between variable values decreases with increasing distance and stabilizes after a certain distance. />The parameters control the scale of the half-variational function, i.e. the rate at which the correlation decreases with increasing distance. />The parameters represent the overall variability between variable values.

specifically, based on the historical interpolation data, an interpolation estimation value at each data point and uncertainty thereof are calculated, such as using statistical methods, such as analysis of variance, root Mean Square Error (RMSE), mean Square Error (MSE), etc., to evaluate the difference between the interpolation result and the historical observation data. Confidence intervals, such as 95% confidence intervals, are calculated to determine the confidence range of the estimate.

Specifically, interpolation is randomly input: interpolation is performed by using the randomly generated input data, and first randomly input interpolation data and second randomly input interpolation data are generated. Semi-variation function: and carrying out semi-variant function interpolation on the first random input interpolation data and the second random input interpolation data according to parameter setting of the semi-variant function. Historical data confidence assessment: and performing confidence evaluation on the first random input interpolation data and the second random input interpolation data by using the historical water and rain condition data/the historical rainfall data to determine final water and rain condition data/rainfall forecast data.

Preferably, step S135 specifically includes:

specifically, a linear interpolation method is used for carrying out time interpolation on the original water rain condition cleaning data and the original rainfall forecast cleaning data. Linear interpolation may perform linear interpolation or extrapolation between data at known time points to estimate the value of the missing time point. For example, if the original data is recorded once per hour, but there is some loss of data at a point in time, linear interpolation may use the data at adjacent points in time to interpolate and calculate an estimate of the point in time at the loss.

specifically, the original water rain condition cleaning data and the original rainfall forecast cleaning data are subjected to spatial distribution interpolation by using a kriging interpolation method, and the interpolation method considers the spatial relationship among data points to estimate the data value of the missing position. First, spatial variation processing is performed: the data is converted into spatially varying data based on the geographic coordinates of the original data points to better estimate the value of the missing location. Then, local spatial trend recognition is performed: the spatial variation data is analyzed, local spatial trends are identified, and the analysis is completed through a statistical method or a spatial covariance function. Next, the kriging interpolation method is used: based on the local spatial trend data and the values of the raw data points, the value of the missing locations is estimated using a kriging interpolation method.

specifically, the first time interpolation data and the second time interpolation data are subjected to weighted fusion to obtain final water and rain condition data and rainfall forecast data. The weights may be determined based on factors such as the confidence level of the data, the time interval, the variance of the kriging interpolation, and the like. For example, if the time interval of the first time-interpolated data is smaller, the reliability is higher, higher weight may be given to the second time-interpolated data, while the weight of the second time-interpolated data is lower, to ensure a better data fusion effect.

in particular, the spatial distances between each of the samples are calculated, and these distances are used to quantify the spatial correlation between different locations, such as using Euclidean distances or other distance metrics to calculate the distances between the samples.

Specifically, to capture local spatial trends, a local region or neighborhood needs to be determined, such as a circular or elliptical neighborhood centered on the point to be estimated (where the radius is determined based on spatially varying data) is selected. In the local neighborhood, spatial trend data of the sample points is calculated by fitting a trend model, such as linear regression, polynomial regression or other model.

Specifically, for the missing position to be estimated, the spatial distance between it and all the known samples is first calculated. Then, the value of the missing position is estimated using the kriging interpolation method based on the spatial distance and the observed value of the known sample point. The kriging interpolation method involves weighting samples in the neighborhood using a weighting function, which is typically related to distance and spatial correlation. The interpolation value is calculated by weighting the observed values of known samples, wherein the higher weighted samples have a greater impact on the estimated value and the lower weighted samples have a lesser impact on the estimated value. Repeating the steps, and performing Kriging interpolation on all the missing positions to obtain an estimated value.

According to the invention, through linear interpolation (the first time interpolation data in the step S135), missing values in a time sequence can be filled, so that the water and rain condition data and rainfall forecast data are ensured to be continuous in time, and the data integrity is maintained. Using spatially distributed kriging interpolation (second time interpolation data in step S135), the missing values can be estimated from the spatial variation of the data. The kriging interpolation method considers the spatial relationship and can better reflect the spatial distribution characteristics of the data. And carrying out weighted fusion on the first time interpolation data and the second time interpolation data, and generating more accurate and reliable water and rain condition data and rainfall forecast data by utilizing the results of different time interpolation. By adopting time interpolation and space distribution Kriging interpolation, continuity of the generated water rain condition data and rainfall forecast data can be ensured, and the model establishment and analysis are facilitated. The interpolation method fills up the missing value in the data, improves the availability of the data, and enables the location and the time point of the missing data to be used for building and analyzing the flood early warning model. Because the interpolation method provides more complete and fine water rain condition data and rainfall forecast data, the accuracy of the flood early warning system is improved, the flood risk is identified in advance, and necessary measures are taken.

Preferably, step S2 is specifically:

specifically, the unit depth per hydrology is a basic parameter in the flood model that represents the runoff flow per unit time per unit area caused by the rainfall over a unit area. The implementation steps are as follows: rainfall data and geographic information data are obtained from the water and rain condition data and are converted into unit hydrologic unit depth data. The unit hydrologic unit depth is estimated by, for example, radar rainfall data, satellite remote sensing data, weather station data, etc., typically in millimeters per hour.

in particular, the region of interest is divided into different spatial units, such as grid units, basin units, or sub-basin units, using Geographic Information System (GIS) tools. And carrying out hydrologic characteristic analysis on each space unit, wherein the hydrologic characteristic analysis comprises information such as gradient, land utilization, soil type and the like, so as to determine hydrologic parameters. And estimating the runoff amount after the unit hydrologic unit depth treatment in each space unit.

Specifically, the boundaries of the investigation region are determined and appropriate boundary conditions, such as quantitative inflow or quantitative outflow conditions, are set. Nearby rivers, lakes or bodies of water are considered to simulate the input and output of a water flow. The effect of the downstream investigation region is taken into account for reflection in the simulation.

In particular, the water flow is simulated using a numerical hydrodynamic model, such as a two-dimensional shallow water equation model or a three-dimensional numerical hydrodynamic model. Based on the water flow dividing data and boundary conditions, time stepping hydrodynamic force calculation is performed to simulate the movement and distribution of water flow. And outputting hydrodynamic data, including information such as water level, flow velocity, flow rate and the like.

Preferably, step S3 is specifically:

in particular, city building distribution data may be obtained from a number of sources, such as Geographic Information Systems (GIS), satellite images, city planning sector data, remote sensing data, and the like. Urban building distribution data typically exists in a vector data format (e.g., shapefile, KML) or a raster data format (e.g., raster image or remote sensing image).

specifically, the artificial building material data including information of building type, roof type, wall type, etc. is obtained from city planning departments, building design files, satellite image analysis, etc. The parameters of the infiltration rate (infiltration rate) or the permeability (permaability) of the building materials are determined according to different building materials, for example, the infiltration rate of different materials such as concrete, tiles, lawns and the like can be different. The infiltration rate of each building is calculated according to the distribution of the building and the material information by using a hydrologic model or formula, and different parameters such as horizontal infiltration rate (horizontal infiltration rate) or vertical infiltration rate (vertical infiltration rate) are adopted.

Specifically, natural soil type data, typically including soil characteristics such as soil texture, porosity, saturated water conductivity, are obtained from soil surveys, soil maps, geological surveys, and the like. The soil types are mapped to corresponding soil parameters, and different types of soil have different infiltration rates and hydrologic characteristics, for example sandy soil typically has a higher permeability to water, while clay soil typically has a poorer permeability to water. The natural infiltration rate of each region is calculated according to the soil type and the soil parameters by using a hydrologic model or formula, and can be quantitative numerical calculation or a parameterized model based on the soil characteristics.

Preferably, step S4 is specifically:

acquiring pipeline drainage data;

in particular, plumbing displacement data of municipal drainage systems are obtained, which are typically monitored and recorded by municipal drainage authorities or water resource authorities. The water displacement data may be time series data, typically including hourly, daily or monthly flow information.

in particular, the plumbing drainage data is integrated with the city building distribution data to build the required input data set for the city hydrologic model. Urban terrain data, including terrain elevation, slope and other terrain information, is obtained from Digital Elevation Models (DEMs) or laser radar scanning and other modes. And marking the drainage points of the pipeline drainage system and the building distribution data to determine the position and attribute information of each drainage point. Urban terrain classification is performed based on terrain data, typically dividing the city into different areas or units for hydrographic simulation.

Specifically, a hydrologic model of the city is established by combining water flow data and city topography division data using a hydrologic model (such as SWMM, HEC-HMS, etc.). Various parameters including a flow calculation method, a confluence model, hydrologic soil parameters and the like are set for the hydrologic model so as to ensure that the model accurately reflects actual conditions. The hydrologic process in the city is simulated by utilizing the hydrologic model, wherein the hydrologic process comprises rainwater runoff, surface runoff, river flow and the like. These simulations may be performed based on time series data.

specifically, a hypotonic model, such as the GR4J model, is selected for estimating the urban hypotonic process. Corresponding parameters including soil water content parameters, infiltration rate parameters and the like are set for the infiltration model so as to carry out infiltration treatment according to urban infiltration rate data. The urban infiltration rate data is combined with hydrologic space-time simulation data, and the actual infiltration process in the urban area is calculated, such as time sequence and space distribution infiltration calculation.

In particular, hydrodynamic models, such as one-dimensional river hydrodynamic model (HEC-RAS), two-dimensional hydrodynamic model (FLO-2D), etc., are selected for simulating the water flow process in the city. Parameters including terrain data, hydrodynamic parameters, boundary conditions, etc. are set for the hydrodynamic model to ensure that the model can accurately simulate the water flow process. Combining the hydrologic space-time infiltration data with a hydrodynamic model, and calculating runoff processes in the city, including flood evolution, flow velocity distribution and the like.

Preferably, step S5 is specifically:

specifically, the model simulates: and simulating rainfall conditions in a preset time period by using the urban surface hydrologic model, wherein the rainfall conditions comprise the steps of inputting rainfall data, simulating hydrologic processes and calculating surface water area conditions. Generating surface water area condition data: surface water volume condition data, including surface water volume peak change data and surface water volume depth change data, are extracted from the output of the hydrologic model reflecting the accumulation of surface water bodies during rainfall events.

Specifically, flood grade classification: the surface water area condition data is classified into different flood grades according to certain criteria and thresholds, for example, different flood grades, such as small flood, medium flood, large flood, etc., can be defined. Flood grade generation: and generating flood grade data based on the classified data, wherein the flood grade data is used for describing the intensity and severity of flood caused by the current rainfall event, such as digital codes, color codes or text descriptions. Flood range generation: by analyzing surface water area condition data and flood grade data, a flood influence range is determined, and space analysis and visualization are performed through GIS (geographic information system) technology and model. Flood early warning: and carrying out flood early warning according to the generated flood grade data and flood range data, wherein the flood early warning can be issued to related departments and the public in an automatic system, an alarm, a notification and other modes so as to take necessary emergency measures.

The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The flood early warning and forecasting method is characterized by comprising the following steps:

step S1: acquiring water rain condition data and rainfall forecast data;

2. The method according to claim 1, wherein step S1 is specifically:

3. The method according to claim 2, wherein step S13 is specifically:

wherein the step of interpolation processing includes the steps of:

4. A method according to claim 3, wherein step S134 is specifically:

5. A method according to claim 3, wherein step S135 is specifically:

6. The method according to claim 1, wherein step S2 is specifically:

7. The method according to claim 1, wherein step S3 is specifically:

8. The method according to claim 1, wherein step S4 is specifically:

acquiring pipeline drainage data;

9. The method according to claim 1, wherein step S5 is specifically:

10. A flood early warning and forecasting system for performing the flood early warning and forecasting method as claimed in claim 1, the flood early warning and forecasting system comprising: