River water regime forecasting method and system based on digital twinning

Abstract

The present application relates to the technical field of water regime forecasting, and particularly relates to a river water regime forecasting method and system based on digital twinning. The method comprises the following steps: obtaining real-time water level data along the river, real-time flow data along the river and real-time meteorological data along the river by arranging sensor nodes along the river and monitoring the river environment and water regime in real time along the river; generating a river water regime flood risk estimation period and a river water regime drought risk estimation period by constructing digital twinning and risk estimation along the river; performing corresponding river water regime flood risk management decision-making work by performing flood risk forecasting processing in the river water regime flood risk estimation period; performing corresponding river water regime drought risk management decision-making work by performing drought risk forecasting processing in the river water regime drought risk estimation period. The present application can realize efficient and accurate water regime prediction and decision support.

Description

Technical Field

[0001] The present invention relates to the technical field of water regime forecasting, and in particular to a river water regime forecasting method and system based on digital twins. Background Art

[0002] Digital twins map physical entities or systems into virtual models in real time, combining real-time data, simulation, and artificial intelligence algorithms to enable dynamic monitoring, prediction, and optimization of physical entities. Digital twin-based systems can simulate and predict the behavior of complex systems, demonstrating significant potential in fields such as hydrological environments and meteorological monitoring. Furthermore, in the field of river water regime forecasting, digital twin technology can acquire multi-source data (such as precipitation, flow rate, water level, and soil moisture) within a river and its basin in real time, combined with efficient simulation models, to construct a virtual twin of the river hydrological system in real time, accurately reflecting the actual conditions within the river basin. Furthermore, digital twin systems can leverage machine learning and big data analytics to update water regime data in real time and adjust water regime prediction models based on changing environmental conditions, improving forecast accuracy and timeliness. However, existing forecasting methods primarily rely on single historical data or fixed models, failing to reflect real-time changes in dynamic factors such as river hydrology, meteorology, and topography. This makes it difficult to adapt to sudden and complex water regime changes, resulting in poor forecast accuracy and timeliness. Summary of the Invention

[0003] Based on this, it is necessary for the present invention to provide a river water condition forecasting method and system based on digital twins to solve at least one of the above technical problems.

[0004] To achieve the above objectives, a river water regime forecasting method based on digital twins includes the following steps:

[0005] Step S1: by deploying sensor nodes along the river channel, and based on the sensor nodes, real-time monitoring of the river environment and water conditions along the river channel is performed to obtain real-time water level data, real-time flow data, and real-time meteorological data along the river channel, wherein the real-time meteorological data along the river channel includes rainfall along the river channel, temperature along the river channel, and wind speed along the river channel; using the real-time water level data, real-time flow data, and real-time meteorological data along the river channel, a digital twin of the river channel is constructed to generate a digital twin model of the environmental water conditions along the river channel;

[0006] Step S2: Based on the rainfall along the river, risk estimation is performed on the corresponding river water period in the digital twin model of the environmental water regime along the river to generate a river water flood risk estimation period and a river water drought risk estimation period;

[0007] Step S3: Obtain the corresponding water level along the river, the flow along the river, and the wind speed along the river in the river regime flood risk estimation period, and perform flood risk prediction processing on the corresponding river environment regime digital twin model based on the water level along the river, the flow along the river, and the wind speed along the river, to generate a river regime flood risk prediction result, to perform corresponding river regime flood risk management decision work;

[0008] Step S4: Obtain the corresponding water level along the river, the flow along the river, and the temperature along the river in the river regime drought risk estimation period, and perform drought risk prediction processing on the corresponding river environment regime digital twin model based on the water level along the river, the flow along the river, and the temperature along the river, to generate a river regime drought risk prediction result, to perform corresponding river regime drought risk management decision work.

[0009] Further, step S4 includes the following steps:

[0010] Step S41: Obtain the water level along the river, the flow along the river, and the temperature along the river in the drought risk period corresponding to the river environment regime digital twin model in the river regime drought risk estimation period;

[0011] Step S42: Perform time series fluctuation cross and lag evaluation analysis on the water level along the river and the flow along the river in the drought risk period, to obtain the time series fluctuation cross interaction and time series fluctuation lag effect between the water level along the river and the flow along the river;

[0012] Step S43: Perform time series fluctuation synchronization analysis on the water level along the river and the flow along the river in the drought risk period based on the time series fluctuation cross interaction and time series fluctuation lag effect between the water level along the river and the flow along the river, to generate a corresponding river hydrological fluctuation response field in the drought risk period;

[0013] Step S44: Perform hydrological evaporation rate evaluation analysis on the corresponding river hydrological fluctuation response field in the drought risk period based on the temperature along the river, to obtain the influence correlation between river hydrological fluctuation and river surface evaporation rate;

[0014] Step S45: Perform drought risk prediction processing on the corresponding river environment regime digital twin model based on the influence correlation between river hydrological fluctuation and river surface evaporation rate, to generate a river regime drought risk prediction result, to perform corresponding river regime drought risk management decision work.

[0015] Further, the application also provides a river water regime prediction system based on digital twinning, used for executing the river water regime prediction method based on digital twinning as described above, and the river water regime prediction system based on digital twinning comprises:

[0016] a river water regime digital twinning modeling module, configured to: arrange sensor nodes along a river, and perform real-time monitoring on the river environment and water regime along the river based on the sensor nodes to obtain real-time water level data along the river, real-time flow data along the river, and real-time meteorological data along the river, wherein the real-time meteorological data along the river comprises rainfall along the river, temperature along the river, and wind speed along the river; and perform digital twinning construction on the river based on the real-time water level data along the river, the real-time flow data along the river, and the real-time meteorological data along the river to generate a digital twinning model of the environment and water regime along the river;

[0017] a river water regime risk estimation time period division module, configured to: perform risk estimation division on a corresponding river water regime time period in the digital twinning model of the environment and water regime along the river based on the rainfall along the river to generate a river water regime flood risk estimation time period and a river water regime drought risk estimation time period;

[0018] a river water regime flood risk prediction module, configured to: acquire corresponding water level along the river, flow along the river, and wind speed along the river in the river water regime flood risk estimation time period, and perform flood risk prediction processing on the corresponding digital twinning model of the environment and water regime along the river based on the water level along the river, the flow along the river, and the wind speed along the river to generate a river water regime flood risk prediction result, so as to execute corresponding river water regime flood risk management decision work;

[0019] a river water regime drought risk prediction module, configured to: acquire corresponding water level along the river, flow along the river, and temperature along the river in the river water regime drought risk estimation time period, and perform drought risk prediction processing on the corresponding digital twinning model of the environment and water regime along the river based on the water level along the river, the flow along the river, and the temperature along the river to generate a river water regime drought risk prediction result, so as to execute corresponding river water regime drought risk management decision work.

[0020] The application has the following beneficial effects:

[0021] 1、The river water regime prediction method based on digital twinning provided in the present application has the beneficial effects that, compared with the prior art, by arranging sensor nodes along the river, key environmental data of the river can be collected in real time, the sensor nodes serve as the front line of data collection and can comprehensively monitor the water level, flow and weather along the river, thereby providing accurate data basis for subsequent water regime analysis and decision support. Through real-time monitoring of the river environment and water regime along the river based on the sensor nodes, the water level sensor can not only automatically collect water level data, but also upload and process data through the Internet of Things and the cloud platform, so as to timely identify the change trend of the water level, the flow sensor can help the relevant departments to timely discover abnormal flow fluctuation and make corresponding emergency warning, and the meteorological monitoring sensor can monitor the weather along the river in real time, so as to obtain important meteorological data including rainfall, temperature and wind speed, thereby reflecting the changes of dynamic factors such as river hydrology, weather and terrain in real time. At the same time, by uploading the real-time data of the water level along the river, the real-time data of the flow along the river and the real-time data of the weather along the river to the cloud platform, and using the real-time data of the water level along the river, the real-time data of the flow along the river and the real-time data of the weather along the river to construct the digital twinning along the river, the digital twinning model of the river can be constructed by using these data, the digital twinning technology can copy the physical and environmental state of the river to the digital platform through virtualization, so that the manager can monitor and simulate the changes of the river in real time, the digital twinning model can simulate the changes of the river level and flow when the flood occurs, predict the pressure change of the reservoir or dam, and improve the accuracy of emergency response, thereby effectively dealing with the sudden and complex water regime changes. Secondly, by dividing the corresponding river water regime period in the river environment water regime digital twinning model based on the rainfall along the river, the corresponding water regime risk period can be effectively identified and estimated, this process makes the water regime monitoring not only stop at general water level and flow observation, but also turn to time and space division based on actual risk, by segmenting the river water regime fluctuation according to flood and drought risk, real-time risk assessment can be provided for various extreme weather events, further optimizing water resource management and disaster prevention and mitigation work, which not only provides fine decision basis for subsequent flood and drought disaster prevention, but also provides more comprehensive and in-depth data support for subsequent management decision, thereby improving the accuracy and timeliness of the prediction result.Then, the river water regime data under the flood risk period is obtained through the digital twin model, which can provide real-time and accurate water level, flow and wind speed and other important parameters. These data provide the basis for subsequent analysis, and the corresponding river environmental water regime digital twin model is processed for flood risk prediction based on the river water level, river flow and river wind speed, to generate the final water regime flood risk prediction result. By considering multiple factors such as water level, flow and wind speed, and through the influence analysis of dynamic response correlation field and micro change factor, more accurate flood risk prediction can be generated. This process is not only data processing, but also in-depth integration and system modeling of various hydrological and meteorological factors, so as to form a flood risk prediction with high predictability. The prediction result will directly affect the river water regime flood risk management decision-making work, helping relevant departments to make scientific decisions before the flood occurs, such as strengthening flood protection facilities, river dredging, evacuation of the public and timely deployment of emergency measures, so as to minimize the loss and impact of the flood. Finally, by using digital twin technology, the hydrological environment along the river can be accurately simulated and predicted, especially in the drought risk period, the changing trend of river water level, flow and temperature can be dynamically reflected, providing solid foundation data for subsequent drought risk assessment. In addition, the corresponding river environmental water regime digital twin model is processed for drought risk prediction based on the river water level, river flow and river temperature, which can realize more accurate drought risk prediction. Drought risk prediction is a key link in water resources management, which can predict the time, range and intensity of drought in advance, provide early warning signals for water resources scheduling and management, and based on this prediction result, water resources management departments can take timely measures such as adjusting reservoir scheduling, increasing water source allocation, and taking water-saving measures. It can also support a wider decision-making process, such as water resource protection area planning, ecological restoration plan and long-term development strategy of water resource use, so as to realize sustainable water resources management.

[0022] 2、The river water regime prediction system based on digital twinning proposed in the present application is composed of a river water regime digital twin modeling module, a river water regime risk estimation period division module, a river water regime flood risk prediction module and a river water regime drought risk prediction module. It can realize any river water regime prediction method based on digital twinning, and can realize the river water regime prediction method based on digital twinning by combining the operations between the computer programs running on each module. The internal structure of the system cooperates with each other, which can greatly reduce the repeated work and manpower investment, and can quickly and effectively provide more accurate and efficient river water regime prediction process based on digital twinning, thereby simplifying the operation process of the river water regime prediction system based on digital twinning. BRIEF DESCRIPTION OF DRAWINGS

[0023] Other features, objects, and advantages of the application will become more apparent from the following detailed description when read in connection with the following drawings:

[0024] Figure 1 A step flow diagram of the river water regime prediction method based on digital twinning of the present application is shown in

[0025] Figure 2 A detailed step flow diagram of step S1 is shown in Figure 1 DETAILED DESCRIPTION

[0026] The technical method of the present application will be described clearly and completely below in conjunction with the drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0027] To achieve the above-mentioned purpose, please refer to Figures 1 to 2 The present application provides a river water regime prediction method based on digital twinning. In the embodiments of the present application, please refer to Figure 1 A step flow diagram of the river water regime prediction method based on digital twinning of the present application is shown in

[0028] Step S1: By arranging sensor nodes along the river, real-time monitoring of the river environment and water regime along the river is carried out based on the sensor nodes to obtain real-time data of water level along the river, real-time data of flow along the river, and real-time data of weather along the river, wherein the real-time data of weather along the river includes rainfall along the river, temperature along the river, and wind speed along the river; digital twinning is constructed along the river by using the real-time data of water level along the river, the real-time data of flow along the river, and the real-time data of weather along the river to generate a digital twinning model of the environment and water regime along the river;

[0029] ​In the embodiment of the present application, by arranging sensor nodes along the river course, the geographical position survey along the river course needs to be carried out first to determine the arrangement position of each sensor node. The water level sensor should be arranged in the representative water area along the river course to ensure that it can reflect the change of the river water level in real time. The flow sensor should be installed in the area with relatively stable water flow to accurately monitor the flow. The weather monitoring sensor should cover the entire river area and be distributed at different geographical heights and climate conditions. The sensor node is connected with the surrounding environment through a wireless communication protocol (such as LoRaWAN, NB-IoT, etc.) to ensure that it can receive and send monitoring data to the central processing unit and connect with the cloud platform through the implementation of Internet of Things technology. First, the hardware module with Internet of Things communication capability, such as Wi-Fi, LoRa, NB-IoT, etc. wireless communication module, needs to be configured in the sensor node. These modules send the real-time collected water level, flow and weather data to the cloud data platform through wireless mode. The water level data along the river course collected by the water level sensor in the Internet of Things network can be uploaded to the cloud in real time. The platform displays the change of the river water level in real time to obtain the real-time data of the water level along the river course. The flow sensor is used to monitor the real-time flow of the corresponding river course. These sensors can measure the speed and flow of water flow by using electromagnetic principle or ultrasonic principle. By monitoring the change of the speed of water flow and the cross-sectional area of water body, the accurate flow data is calculated to obtain the real-time flow data along the river course. At the same time, the weather monitoring sensor is used to monitor the real-time environment and weather of the river course. The weather monitoring sensor should adopt multiple weather parameter detection modules to monitor rainfall, temperature, humidity and wind speed data respectively. The weather sensor node is usually configured with high-precision rainfall gauge, temperature and humidity sensor and anemometer and other equipment. These sensors collect weather data regularly and send the data to the cloud platform through wireless communication to obtain real-time weather data along the river course, including rainfall along the river course, temperature along the river course and wind speed along the river course. Then, after uploading the real-time water level, flow and weather data along the river course obtained by real-time monitoring to the cloud platform, the platform will generate a water regime digital twin model along the river course according to the real-time data obtained, through digital twin technology. The digital twin technology first needs to construct a physical model according to the data collected by the sensor and add real-time data source in the model to form a closed loop between each node and data. The cloud platform will analyze the multi-dimensional data through machine learning and data modeling technology according to the real-time water level, flow, weather data and historical data to display the real-time digital twin of the water regime along the river course. The platform uses these data to simulate the change of water flow in the river through algorithm to reflect the rise and fall of water level and the fluctuation of flow in real time. According to the output of the model, the platform will generate visual reports, graphs and prediction results to finally generate an environmental water regime digital twin model along the river course.

[0030] Step S2: Based on the rainfall along the river, the corresponding river water regime period in the river along the environment water regime digital twin model is risk estimated and divided to generate the river water regime flood risk estimation period and the river water regime drought risk estimation period;

[0031] In the embodiment of the present application, by collecting rainfall data along the river, these data are usually obtained by meteorological stations or remote sensing technology, and need to cover a certain time range, usually monthly or annual data, then based on the river water regime data in the digital twin model, especially the water level, flow and storage and other hydrological data synchronized with rainfall, spatio-temporal analysis is carried out to ensure the consistency of rainfall and water regime data in time, the correlation analysis method (such as Pearson correlation coefficient, Spearman rank correlation, etc.) is used to calculate the correlation between rainfall and water regime fluctuation (flow, storage, etc.), and then the spatio-temporal correlation between rainfall and water regime fluctuation is revealed, and further water regime fluctuation time series feature analysis is carried out according to the previously obtained spatio-temporal correlation between rainfall and water regime fluctuation, first, the rainfall and water regime fluctuation data of different water regime periods (such as drought period, rainy season, etc.) along the river are decomposed in time series, and the periodic and aperiodic fluctuation characteristics are extracted by wavelet analysis, Fourier transform and other methods, the periodic fluctuation characteristics can be analyzed by periodic chart or fluctuation amplitude frequency characteristics extraction, and the aperiodic fluctuation characteristics can be obtained by detrending or difference method, then these time series feature data are integrated into feature data set of rainfall, flow fluctuation and storage change, at the same time, a suitable water regime fluctuation risk judgment calculation formula is constituted by combining rainfall, rainfall rate, river flow, river flow maximum threshold, river storage change, river storage change maximum threshold and related parameters to carry out risk judgment and evaluation calculation on the corresponding river water regime period in the river along the environment water regime digital twin model, so as to judge whether the period enters the flood risk stage or the drought risk stage, so as to obtain the flood and drought fluctuation risk judgment parameters, then, according to the aforementioned flood and drought fluctuation risk judgment parameters, each period is segmented according to the extreme risk, in this process, first, the corresponding river water regime fluctuation in the river water regime period is compared according to the water regime fluctuation data and the flood risk and drought risk judgment parameters, if the water regime fluctuation (such as flow fluctuation, storage change, etc.) of a period exceeds the predetermined flood risk judgment range, the period is marked as a flood risk estimation period; on the contrary, if the water regime fluctuation is within the judgment range of drought risk, the period is judged as a drought risk estimation period, in order to ensure the accuracy of the judgment, dynamic threshold judgment method can be used to adjust the threshold interval in real time according to the actual situation, this step can be realized by conditional judgment logic in programming language such as Python combined with risk judgment formula, and the final output is a clear time period division which specifically identifies the risk estimation period of flood or drought.

[0032] Step S3: Obtain the corresponding water level along the river, the flow along the river and the wind speed along the river in the river regime flood risk prediction period, and based on the water level along the river, the flow along the river and the wind speed along the river, the corresponding environmental regime digital twin model along the river is processed for flood risk prediction, and the river regime flood risk prediction result is generated to execute the corresponding river regime flood risk management decision work;

[0033] In the embodiment of the present application, by obtaining the river water regime digital twin model along the river course during the flood risk prediction period, the model is ensured to have the function of reflecting the actual hydrological and hydraulic environment. In the model, the river water level, flow and wind speed and other data are taken as key parameters, the real-time water level, flow, wind speed and other data of the river course are regularly obtained, and the historical river water level fluctuation data and climate model are obtained as the basis for analysis. The evolution of the river water level is analyzed by the historical observation data and the climate simulation data. Firstly, the river water level fluctuation data of the past years are collected and analyzed. These data can be obtained from the meteorological bureau, the hydrological department and the local weather station. Then, the future climate change trend is predicted by using the climate model, especially the change of precipitation, temperature change and other factors. Combined with these historical data and climate model, a water level evolution analysis model is established to calculate the water level change trend during the flood risk period and generate the water level fluctuation field during this period. At the same time, the river flow during the flood risk period is monitored in real time to obtain the flow data and further analyze the correlation between the flow and the upstream inflow, the midstream precipitation and the downstream outflow. Firstly, multiple flow monitoring points are set up, and flow meters and other equipment are used to obtain the flow data of each part of the river in real time. Secondly, the remote sensing monitoring and the data of the hydrological and meteorological station are used to estimate the upstream inflow and the midstream precipitation. The flow and water level evolution coupling analysis is carried out by inputting these data into the calculation model. The response function between the flow change and the water level fluctuation is established to generate the dynamic response correlation field combined with the historical evolution trend of the water level. The influence factor evaluation analysis based on wind speed data is carried out. Firstly, the actual wind speed data is collected along the river course by wind speed monitoring instruments. The wind speed sensor can be installed at different positions of the river course, especially in the areas susceptible to wind speed changes. Combined with the meteorological data, the micro influence of wind speed on the change of river water level and flow is evaluated. The influence degree of wind speed on the water surface and flow is determined by numerical simulation and fluid mechanics analysis. For example, wind speed can cause disturbance of the water surface, thereby changing the distribution of water flow and water level change. The fluid dynamics model (such as CFD model) is used to add the wind speed as the influence factor into the change model of river water level and flow, so as to obtain the quantitative influence factor of wind speed on the change of water level and flow. Then, based on the micro change influence factor generated in the foregoing steps during the flood risk period, the river water regime digital twin model is used for flood risk prediction. This process firstly combines the water level, flow, wind speed and other data obtained in the previous steps to establish a dynamic response model in the water regime digital twin model by fusing all influence factors. The model can predict the change of water level and flow according to the real-time monitoring data and historical trends, and adjust according to different flood situations. Specifically, the real-time water level, flow and wind speed data are input, numerical simulation method is used for prediction processing, and the prediction results of river water level and flow are output.The forecast result is used to determine the flood risk level, and a decision support is provided for the river management department according to the flood risk level, so as to guide the water regime management and disaster prevention and mitigation work, and finally execute the corresponding river water regime flood risk management decision work.

[0034] Step S4: Obtain the corresponding water level along the river, the flow along the river and the temperature along the river in the river water regime drought risk prediction period, and perform drought risk prediction processing on the corresponding environment water regime digital twin model along the river based on the water level along the river, the flow along the river and the temperature along the river, to generate a river water regime drought risk prediction result, so as to execute the corresponding river water regime drought risk management decision work.

[0035] In the embodiment of the present application, by using the water regime model based on digital twin technology to obtain the water level, flow and temperature data along the river channel in the corresponding drought risk period range during the drought risk prediction period, a three-dimensional digital twin model of the river channel is first constructed, including the geometric shape of the river channel, historical hydrological data and real-time sensor data, combined with the weather prediction model and real-time observation data, the water level, flow and temperature data of the river channel are collected in real time through remote sensing technology and Internet of Things sensors (such as water level sensors, flow meters, temperature sensors, etc.), and after obtaining the water level and flow data along the river channel, the time series fluctuation cross and lag effect are evaluated and analyzed, the specific method is to use time series analysis technology to analyze the relationship between river water level and flow in detail, first, through statistical analysis methods such as correlation analysis, cointegration analysis, etc., the time series fluctuation cross effect between water level and flow is identified, and through cross-correlation function (CCF) analysis, the correlation between water level and flow at different time points is calculated to find out their interaction relationship, secondly, through the use of lag effect analysis, the influence delay time of water level change on flow change is evaluated, using lag regression model or Granger causality test method, the time lag effect of water level change on flow is quantified, at the same time, after completing the time series fluctuation cross and lag effect analysis of water level and flow, based on the analysis results, the time series fluctuation synchronization analysis of water level and flow is carried out, the synchronization analysis technology such as waveform correlation and phase synchronization analysis is used to quantify the synchronization and synchronization change degree of water level and flow change, on this basis, the hydrological fluctuation response field along the river channel is generated through simulation model, and in the drought risk period, the relationship between temperature and hydrological fluctuation response field along the river channel is evaluated to further analyze the hydrological evaporation rate, first, based on the temperature data along the river channel, combined with the hydrological fluctuation response field, the evaporation calculation model (such as Penman-Monteith model or Hargreaves model) is used to evaluate the evaporation rate under different temperature conditions, through the correlation between temperature, humidity, wind speed and other environmental factors and water level, flow fluctuation, the influence relationship between hydrological fluctuation and evaporation rate is quantified, for different river channel sections, according to the actual measured water surface temperature and environmental data, the corresponding evaporation rate is calculated, and through regression analysis, correlation analysis and other methods, the influence correlation relationship between water surface evaporation rate and water level fluctuation is obtained.Then, by associating the influence between the previously obtained hydrological fluctuations and evaporation rates, combined with historical hydrological data and real-time monitoring data, the water regime along the river is drought risk forecasted, and the drought risk of the water regime is simulated through the digital twin model. Based on the obtained environmental data (such as water level, flow, temperature, etc.), real-time analysis and prediction are carried out, drought risk assessment is carried out using the model, the model predicts the future changes of river water level and flow according to the hydrological fluctuation response field and water surface evaporation rate, simulates the range, intensity and duration of drought risk possible influence, and outputs the risk prediction result. The result provides decision support for river water regime management. Especially when the drought risk is high, decision makers can arrange water resources dispatching, drought mitigation measures, etc. according to the prediction result, take relevant water regime management and emergency measures to ensure the rational use and protection of river water resources, and finally implement the corresponding river water regime drought risk management decision work.

[0036] Further, as an embodiment of the present application, referring to Figure 2 the detailed step flowchart of step S1 in the embodiment, step S1 includes the following steps: Figure 1

[0037] Step S11: arranging sensor nodes along the river, wherein the sensor nodes include water level sensors, flow sensors and weather monitoring sensors;

[0038] In the embodiment of the present application, by arranging sensor nodes along the river, the geographical position survey along the river needs to be carried out first to determine the arrangement position of each sensor node. The water level sensor should be arranged in the representative water area along the river to ensure that it can reflect the change of the river water level in real time. The flow sensor should be installed in the area with relatively stable water flow to accurately monitor the flow. The weather monitoring sensor should cover the entire river area and be distributed at different geographical heights and climate conditions. The sensor nodes are connected with the surrounding environment through wireless communication protocols (such as LoRaWAN, NB-IoT, etc.) to ensure that they can receive and send monitoring data to the central processing unit. The power system of each sensor node should have sufficient endurance, especially for some remote sensor nodes, which usually need to use solar energy or other environmentally friendly energy to ensure their long-term stable operation.

[0039] Step S12: based on the sensor nodes, using Internet of Things technology to connect with the cloud platform for monitoring network connection to generate a water regime monitoring Internet of Things network along the river; using the water level sensors in the water regime monitoring Internet of Things network along the river to monitor the water level along the river in real time to obtain real-time water level data along the river;

[0040] ​In an embodiment of the present invention, by realizing the connection between the Internet of Things technology and the cloud platform, first, the sensor nodes need to be configured with hardware modules with Internet of Things communication capabilities, such as Wi-Fi, LoRa, NB-IoT and other wireless communication modules. These modules will send the real-time collected water level, flow and meteorological data to the cloud data platform wirelessly. When realizing the Internet of Things network connection, the cloud platform needs to have high-performance data reception and processing capabilities, and be able to receive data uploaded by each sensor node through multiple protocols (such as MQTT, HTTP, etc.). At the same time, the platform needs to set corresponding access permissions and security verification mechanisms to ensure the security of data transmission, thereby connecting to generate the Internet of Things network for water condition monitoring along the river. The water level data along the river collected by the water level sensors in the Internet of Things network for water condition monitoring along the river can be uploaded to the cloud in real time. The platform displays the changes in the river water level in real time, and stores and analyzes the uploaded data, and finally obtains the real-time water level data along the river.

[0041] Step S13: using the flow sensors in the water regime monitoring IoT network along the river to monitor the flow along the river in real time, so as to obtain real-time data on the flow along the river;

[0042] In an embodiment of the present invention, flow sensors are used to monitor the water flow along the corresponding river in real time. The flow sensors should be deployed at suitable river intersections or areas with large flow rates and rapid changes. These sensors can use electromagnetic principles or ultrasonic principles to measure the speed and flow of water flow, and calculate accurate flow data by monitoring the changes in the speed of water flow and the cross-sectional area of ​​the water body. Under the Internet of Things network connection, the sensors upload the flow data to the cloud platform at regular intervals. The cloud platform automatically processes this data through algorithms, calculates the actual flow of the river at the current moment, and can perform flow trend analysis based on historical data, ultimately obtaining real-time flow data along the river.

[0043] Step S14: using meteorological monitoring sensors within the river water regime monitoring IoT network to monitor the river's weather in real time, thereby obtaining real-time meteorological data along the river, including rainfall, temperature, and wind speed along the river;

[0044] In the embodiment of the present application, by using the meteorological monitoring sensor to monitor the environmental meteorology along the river course in real time, the meteorological monitoring sensor should adopt multiple meteorological parameter detection modules for monitoring rainfall, temperature, humidity and wind speed data, etc. The meteorological sensor node is usually configured with high-precision rainfall gauge, temperature and humidity sensor and anemometer and other equipment. These sensors collect meteorological data regularly and send the data to the cloud platform through wireless communication. Rainfall data can be collected and uploaded in real time by the rain gauge to help monitor the precipitation of the river basin. Temperature and wind speed data are collected in real time by integrated sensors. In the cloud platform, these meteorological data are integrated and analyzed with water level data and flow data to obtain real-time meteorological data along the river course, including rainfall along the river course, temperature along the river course and wind speed along the river course.

[0045] Step S15: uploading the real-time water level data along the river course, the real-time flow data along the river course and the real-time meteorological data along the river course to the cloud platform, and using the real-time water level data along the river course, the real-time flow data along the river course and the real-time meteorological data along the river course to construct the digital twin along the river course to generate the environmental water regime digital twin model along the river course.

[0046] In the embodiment of the present application, after uploading the previously monitored real-time water level, flow and meteorological data along the river course to the cloud platform, the platform will generate a water regime digital twin model along the river course according to the real-time data obtained by the digital twin technology. The digital twin technology first needs to construct a physical model according to the data collected by the sensor, and add a real-time data source in the model, so that a closed loop is formed between each node and data. The cloud platform will analyze the real-time water level, flow, meteorological data and historical data through machine learning and data modeling technology to perform multi-dimensional data analysis and real-time digital twin display of the water regime along the river course. The platform uses these data to simulate the changes of the water flow in the river course through an algorithm, and reflects the rise and fall of the water level and the fluctuation of the flow in real time, thereby providing basic data support for future water regime prediction and flood warning. According to the output of the model, visual reports, graphs and prediction results are generated, and finally the environmental water regime digital twin model along the river course is generated.

[0047] Further, the digital twin construction along the river course using the real-time water level data along the river course, the real-time flow data along the river course and the real-time meteorological data along the river course in step S15 includes the following steps:

[0048] Performing spatio-temporal correlation analysis on the real-time water level data along the river course and the real-time flow data along the river course to obtain the spatio-temporal dynamic correlation between the water level change and the flow fluctuation along the river course;

[0049] In the embodiment of the present application, the water level data and flow data are obtained in real time by deploying water level and flow monitoring devices along the river channel, which include buoys, water level sensors and flow rate sensors, etc., and the data are transmitted to the data center for processing through the sensor network, and the real-time water level and flow data collected are analyzed by using the spatio-temporal data analysis method, such as the dynamic time warping (DTW) algorithm and the spatio-temporal correlation model, to reveal the correlation between the water level and the flow, and the purpose of the spatio-temporal correlation analysis is to identify the regularity of the water level fluctuation and the flow change between different time and space points, analyze the dynamic relationship thereof at different time periods and geographical positions, for example, if the rise of the water level is highly correlated with the increase of the flow in a certain period of time, the corresponding spatio-temporal correlation pattern is automatically captured and generated to reflect the synchronous change trend of the two, and finally the spatio-temporal dynamic correlation between the water level change and the flow fluctuation along the river channel is obtained.

[0050] Preferably, the corresponding geographical information coordinate data along the river channel are obtained, and the real-time water level data along the river channel and the real-time flow data along the river channel are mapped in the spatio-temporal correlation of the water regime based on the spatio-temporal dynamic correlation between the water level change and the flow fluctuation along the river channel and in combination with the geographical information coordinate data, to generate a spatio-temporal correlation feature model of the water regime along the river channel.

[0051] In the embodiment of the present application, the geographical coordinate data along the river channel are obtained by using the geographic information system (GIS), which are obtained through remote sensing images, topographic survey and existing geographic databases to ensure the accuracy of the coordinate data, and the spatio-temporal mapping of the water level and flow data is performed by using the spatial analysis tool (such as the spatial autocorrelation analysis and the Kriging interpolation method) in combination with the spatio-temporal dynamic correlation of the water level and flow along the river channel, specifically, the real-time water level and flow data are combined with the geographical information coordinate data by using the spatial interpolation technology based on the above spatio-temporal correlation analysis result, to realize the spatio-temporal correlation mapping of the water regime data, and through this process, the water level change and flow fluctuation along the river channel are converted into the spatio-temporal distribution map related to the geographical coordinates, to form a water regime data model with spatio-temporal dynamic characteristics, which can accurately reflect the change rule of the water level and flow at different geographical positions, and finally generate a spatio-temporal correlation feature model of the water regime along the river channel.

[0052] Preferably, the spatio-temporal correlation feature model of the water regime along the river channel is analyzed by using the meteorological multi-source data fusion based on the real-time meteorological data along the river channel, to generate an environmental water regime time series model along the river channel.

[0053] In the embodiment of the present application, the spatial and temporal correlation characteristic model of the water regime along the river is analyzed by using meteorological data (such as precipitation, air temperature, wind speed, humidity, etc.) and meteorological multi-source data fusion. The meteorological data is obtained in real time through meteorological stations, satellite remote sensing and meteorological prediction systems. When these meteorological data are fused with the water regime data along the river, a data fusion algorithm (such as Kalman filtering algorithm, multi-sensor fusion technology, etc.) is used to accurately pair data from different sources. The fused data can eliminate errors between different data sources and improve the accuracy of the forecast. In specific implementation, the correlation between meteorological factors and changes in river water level and flow is used to construct a water regime time series model. The time dependence of water level and flow changes is identified through analysis of historical data, and the model is dynamically updated based on real-time input of meteorological data. The model can ultimately reflect the trend of changes in the water regime along the river, further improving the accuracy and timeliness of the water regime forecast, and ultimately generating a river environment water regime time series model.

[0054] Preferably, the digital terrain data corresponding to the river along the river is obtained, including riverbed slope and shoreline contour, and the digital terrain data corresponding to the river along the river is analyzed to generate river basin terrain water flow simulation data.

[0055] In the embodiment of the present application, detailed terrain data including riverbed slope, shoreline contour, watershed range, ground elevation, etc. are obtained by digital terrain mapping along the river. These data are usually obtained through remote sensing technology, laser radar (LiDAR) scanning and terrain measurement equipment (such as total station, differential GPS, etc.). After obtaining the digital terrain data, a watershed hydrological model (such as SWAT model, HEC-RAS model, etc.) is used for terrain simulation analysis to simulate the water flow in the river basin. In specific implementation, first, a digital terrain model is established according to the terrain data, and then the movement law, flow rate and flow of the river flow are introduced into the watershed flow simulation system to simulate the distribution and flow of the water flow under different slope and terrain conditions. Through this simulation analysis, the water flow simulation data of the river basin are generated to reveal the water flow change law at different positions in the watershed. These simulation data will provide the necessary spatial basis for the construction of the environmental water regime digital twin model in the subsequent step, and ultimately generate the river basin terrain water flow simulation data along the river.

[0056] Preferably, the river environment water regime time series model is digitally twinned and fused based on the river basin terrain water flow simulation data along the river to generate a river environment water regime digital twin model.

[0057] In the embodiment of the present application, by fusing the river basin terrain water flow simulation data with the previously generated environmental water regime time series model based on digital twinning, in specific implementation, the digital twinning technology is used to connect the hydrological data (such as water level, flow, meteorological data, etc.) in the real world with the water flow simulation data in the virtual world, and dynamic updating is carried out through a bidirectional feedback mechanism. In this process, first, the terrain simulation data is mapped with each item of hydrological parameter in the time series model to ensure that the virtual model can accurately reflect the water regime dynamics of the real river channel. Secondly, the water regime data in the time series model is trained and optimized through big data analysis and machine learning algorithms (such as deep learning model, neural network, etc.) to improve the prediction ability of the model, and a complete river channel along the environmental water regime digital twinning model is constructed. This model can reflect the changes of hydrological parameters such as river water level and flow in real time, while considering environmental factors such as weather and terrain, realize accurate water regime prediction and dynamic monitoring, and finally generate a river channel along the environmental water regime digital twinning model.

[0058] Further, step S2 includes the following steps:

[0059] Step S21: Based on the rainfall along the river channel, the corresponding river water regime time period in the river channel along the environmental water regime digital twinning model is analyzed for water regime space-time fluctuation correlation, and the space-time fluctuation correlation relationship between the rainfall along the river channel and the water regime fluctuation is obtained.

[0060] In the embodiment of the present application, by collecting rainfall data along the river channel, these data are usually obtained by meteorological stations or remote sensing technology, and need to cover a certain time range, usually monthly or annual data. Next, based on the river water regime data in the digital twinning model, especially the water level, flow and water storage and other hydrological data synchronized with rainfall, space-time analysis is carried out. First, through data cleaning, abnormal values and missing data are eliminated to ensure the consistency of rainfall and water regime data in time. Then, correlation analysis method (such as Pearson correlation coefficient, Spearman rank correlation, etc.) is used to calculate the correlation between rainfall and water regime fluctuation (flow, water storage, etc.), and then the change rule of rainfall and water regime fluctuation in space-time is revealed. This analysis can be realized through statistical functions in MATLAB or Python data analysis tools or special hydrological analysis toolkits, and the space-time fluctuation correlation relationship between rainfall and water regime fluctuation is obtained. Finally, the space-time fluctuation correlation relationship between rainfall along the river channel and water regime fluctuation is obtained.

[0061] Step S22: Based on the spatio-temporal fluctuation correlation relationship between the rainfall along the river channel and the water regime fluctuation, the water regime fluctuation time sequence characteristic analysis is performed on the corresponding river water regime period in the river channel along the line of rainfall and the environmental water regime digital twin model, and the rainfall-water regime fluctuation time sequence characteristic data set is obtained, which includes the periodic fluctuation and non-periodic fluctuation characteristics of rainfall, flow fluctuation and storage change corresponding to the period;

[0062] In the embodiment of the present application, by further analyzing the time sequence characteristics of water regime fluctuation according to the previously obtained spatio-temporal correlation relationship between rainfall and water regime fluctuation, firstly, the rainfall and water regime fluctuation data of different water regime periods (such as dry season, rainy season, etc.) along the river channel are decomposed in time sequence, and the periodic and non-periodic fluctuation characteristics are extracted by using wavelet analysis, Fourier transform and other methods. The periodic fluctuation characteristics can be analyzed by periodic chart or fluctuation amplitude frequency characteristics, and the non-periodic fluctuation characteristics can be obtained by detrended or difference method. Next, these time sequence characteristic data are integrated into the characteristic data set of rainfall, flow fluctuation and storage change, forming a comprehensive time sequence characteristic data set, including the frequency, amplitude, duration, trend and other information of the fluctuation. These time sequence characteristic data can be operated and arranged by using the pandas library in Python, and the mathematical model can be calculated by using the NumPy library, so as to form a time sequence characteristic data set which can be used for further analysis, and finally the river channel along the line of rainfall-water regime fluctuation time sequence characteristic data set is obtained, which includes the periodic fluctuation and non-periodic fluctuation characteristics corresponding to the rainfall, flow fluctuation and storage change.

[0063] Step S23: Based on the rainfall-water regime fluctuation time sequence characteristic data set, the risk judgment calculation formula of water regime fluctuation is used to calculate the risk judgment evaluation of the corresponding river water regime period in the river channel along the line of environmental water regime digital twin model, so as to obtain the river channel along the line of water regime fluctuation risk judgment parameter, which includes the river channel along the line of flood fluctuation risk judgment parameter and the river channel along the line of drought fluctuation risk judgment parameter;

[0064] In the embodiment of the present application, by combining the initial time point, the ending time point, the time variable parameter, the rainfall, the rainfall change rate, the rainfall water regime fluctuation risk weight coefficient, the river flow, the river flow maximum threshold value, the river water storage change, the river water storage change maximum threshold value, the water storage change water regime fluctuation risk weight coefficient, the river flow water regime fluctuation risk weight coefficient, the total number of flood tide events, the average value of river flow of flood tide events, the total number of drought events, the standardized drought evaporation index, the average value of evaporation index of drought events, and the related parameters corresponding to the river water regime period in the river along the environmental water regime digital twin model, a suitable water regime fluctuation risk judgment calculation formula is formed to perform risk judgment and evaluation calculation on the corresponding river water regime period in the river along the environmental water regime digital twin model, so as to judge whether the period enters the flood risk stage or the drought risk stage, and finally obtain the river along the water regime fluctuation risk judgment parameter, which includes the river along the flood fluctuation risk judgment parameter and the river along the drought fluctuation risk judgment parameter.

[0065] Step S24: based on the river along the water regime fluctuation risk judgment parameter, the flood and drought extreme risk segmentation of the corresponding river water regime period in the river along the environmental water regime digital twin model is performed, if the river water regime fluctuation in the corresponding river water regime period is within the range corresponding to the river along the flood fluctuation risk judgment parameter, the river water regime period is determined as the river water regime flood risk estimation period; if the river water regime fluctuation in the corresponding river water regime period is within the range corresponding to the river along the drought fluctuation risk judgment parameter, the river water regime period is determined as the river water regime drought risk estimation period.

[0066] In the embodiment of the present application, according to the aforementioned flood and drought fluctuation risk judgment parameters, the extreme risk segmentation of each period is performed, in the process, first, according to the water regime fluctuation data and the flood risk and drought risk judgment parameters, the corresponding river water regime fluctuation in the river water regime period is compared, if the water regime fluctuation (such as flow fluctuation, water storage change, etc.) of a period exceeds the predetermined flood risk judgment range, the period is marked as the flood risk estimation period; on the contrary, if the water regime fluctuation is within the judgment range of the drought risk, the period is judged as the drought risk estimation period, in order to ensure the accuracy of the judgment, the dynamic threshold judgment method can be used, the threshold interval is adjusted in real time according to the actual situation, this step can be realized by the conditional judgment logic in the programming language such as Python combined with the risk judgment formula, and the final output is a clear time period division, which specifically identifies the risk estimation period of the flood or drought.

[0067] Further, the water regime fluctuation risk judgment calculation formula in step S23 is specifically:

[0068]

[0069] In the formula, RH R is a river flood fluctuation risk determination parameter along the river course G t0 is an initial time point corresponding to a river regime period, t1 is an ending time point corresponding to the river regime period, t is a time variable parameter, P(t) is a rainfall at time t, is a rainfall change rate, a1 is a rainfall regime fluctuation risk weight coefficient, Q(t) is a river flow at time t, Q max is a maximum threshold value of the river flow, H(t) is a river storage change amount at time t, H max is a maximum threshold value of the river storage change, a2 is a storage change regime fluctuation risk weight coefficient, exp is an exponential function, β is a river flow regime fluctuation risk weight coefficient, n is a total number of flood events, i is an item index of the flood event, Q i is a river flow of the i-th flood event, Q avg is an average value of the river flow of the flood event, m is a total number of drought events, j is an item index of the drought event, S j is a standardized drought evaporation index of the j-th drought event, S avg is an average value of the evaporation index of the drought event, η1 is a correction coefficient of the river flood fluctuation risk determination parameter, η2 is a correction coefficient of the river drought fluctuation risk determination parameter.

[0070] The application obtains a regime fluctuation risk determination calculation formula by using a specific mathematical model and verification, which is used for risk determination and evaluation calculation of a corresponding river regime period in an environmental regime digital twin model along a river course. The change of rainfall has a direct impact on the river regime fluctuation, especially the rainfall change in a short time (i.e. ) can reflect the rainfall intensity and change trend, by introducing the rainfall change rate as the influencing factor of the regime fluctuation, the risk of heavy rain or sudden rainfall event can be captured in time, especially in the prediction of flood, the sharp increase of rainfall is an important warning signal before the occurrence of flood or drought event, and rapid identification of the degree of rainfall change helps to predict the possibility of flood or drought in advance. This item can quantify the gap between the current flow and the maximum flow, and assess the corresponding flood or drought risk by weighting. Larger flow gap means higher risk. The change in storage reflects the water storage situation of the river channel. Excessive storage will lead to the risk of dam break or flood overflow, while insufficient storage will cause drought. The change of storage relative to the maximum storage threshold provides a standard for measuring the water storage situation of the reservoir or river channel. By monitoring the change of storage, it can identify which period is at risk of flood due to excessive storage or which period is at risk of drought due to insufficient storage. By introducing the statistical characteristics of flow fluctuation This formula considers the total number of rising tide events and their fluctuation characteristics. By calculating the fluctuation of flow, it can determine whether the water regime fluctuation of the river is abnormal. The degree of flow fluctuation directly affects the prediction accuracy of water regime. Excessive fluctuation usually means an increase in risk. By analyzing the flow of past rising tide events, it can assess whether future water regime is likely to have abnormal fluctuations. Irregular fluctuations in flow are usually related to sudden flood events, so this item helps to identify and prevent flood risks in advance. In addition, the standardized analysis of drought events The standardized drought evaporation index helps to quantify the severity of drought. By standardizing the analysis of historical drought events, it can identify whether the current water regime is in a high-risk stage of drought, especially in the case of small flow and scarce precipitation. The standardized drought index helps to unify the drought assessment standards of different regions and periods, making it possible to compare drought fluctuation risks across time and region, and providing support for accurate drought prediction. The introduction of risk correction coefficients is used to further adjust the risk in the formula. These coefficients take into account other environmental factors or special conditions that affect water regime fluctuations. The introduction of these correction coefficients makes the calculation formula more flexible, allowing it to adapt to different regions and environmental changes, enhancing the accuracy and adaptability of risk assessment. This formula integrates the water regime fluctuations at each time point, allowing for a comprehensive and systematic consideration of the temporal and spatial changes in water regime fluctuations, thus obtaining a risk assessment for a continuous period of time. The integration of time can avoid the impact of sudden fluctuations at a single time point on risk assessment, improving the accuracy of prediction. In summary, this formula fully considers the river flood fluctuation risk determination parameter R H , the river drought fluctuation risk determination parameter R G , the initial time point t0 corresponding to the river water regime period, the end time point t1 corresponding to the river water regime period, the time variable parameter t, the rainfall P(t) at time t, the rainfall change rate The rainfall water regime fluctuation risk weight coefficient α1, the river flow Q(t) at time t, the maximum threshold of river flow Q max The change of river storage H(t) at time t, the maximum threshold of river storage H max, the water regime fluctuation risk weight coefficient of the change of water storage α2, the exponential function exp, the water regime fluctuation risk weight coefficient of the river flow β, the total number of flood events n, the index i of the flood event, the river flow Q of the i-th flood event i , the average value of the river flow Q of the flood event avg , the total number of drought events m, the index j of the drought event, the standardized drought evaporation index S of the j-th drought event j , the average value of the evaporation index S of the drought event avg , the correction coefficient η1 of the flood fluctuation risk determination parameter along the river, the correction coefficient η2 of the drought fluctuation risk determination parameter along the river, wherein the initial time point t0 corresponding to the river regime period, the end time point t1 corresponding to the river regime period, the time variable parameter t, the rainfall P(t) at the time t, the rainfall change rate The rainfall water regime fluctuation risk weight coefficient α1, the river flow Q(t) at the time t, the maximum threshold value Q of the river flow max , the change of the river storage H(t) at the time t, the maximum threshold value H of the change of the river storage max , the water regime fluctuation risk weight coefficient of the change of water storage α2, the exponential function exp, the water regime fluctuation risk weight coefficient of the river flow β, the total number of flood events n, the index i of the flood event, the river flow Q of the i-th flood event i and the average value of the river flow Q of the flood event avg constitute a function relationship of a flood fluctuation risk determination parameter R H along the river

[0071]

[0072] Also by combining the initial time point t0 corresponding to the river regime period, the end time point t1 corresponding to the river regime period, the time variable parameter t, the rainfall P(t) at the time t, the rainfall change rate The rainfall water regime fluctuation risk weight coefficient α1, the river flow Q(t) at the time t, the maximum threshold value Q of the river flow max , the change of the river storage H(t) at the time t, the maximum threshold value H of the change of the river storage max , the water regime fluctuation risk weight coefficient of the change of water storage α2, the exponential function exp, the water regime fluctuation risk weight coefficient of the river flow β, the total number of drought events m, the index j of the drought event, the standardized drought evaporation index S of the j-th drought event i and the average value of the evaporation index S of the drought event avg constitute a function relationship of a drought fluctuation risk determination parameter R G along the river The formula can realize the risk judgment and evaluation calculation process of the corresponding river water regime period in the river environment water regime digital twin model, and through the introduction of the correction coefficient η1 of the river flood fluctuation risk judgment parameter and the correction coefficient η2 of the river drought fluctuation risk judgment parameter, the accuracy and applicability of the water regime fluctuation risk judgment calculation formula can be improved.

[0073] Further, step S3 comprises the following steps:

[0074] Step S31: obtaining the river water level, river flow and river wind speed along the river in the flood risk period corresponding to the river environment water regime digital twin model in the river water regime flood risk estimation period;

[0075] In the embodiment of the present application, by obtaining the river water regime digital twin model in the flood risk estimation period, the model can reflect the actual hydrological and hydraulic environment, and in the model, the river water level, flow and wind speed are used as key parameters. In actual operation, the real-time water level, flow, wind speed and other data of the river are obtained regularly by using remote sensing technology and the data acquisition system of the hydrological monitoring station, and combined with environmental parameters such as precipitation and soil moisture, a high-precision water regime digital twin model is constructed. The model simulates and predicts the corresponding water level change, flow fluctuation and wind speed influence in the flood risk period by integrating these data. This process uses a high-performance computing platform to quickly simulate and iteratively update the water regime model, ensuring that the simulation results accurately reflect the current hydrological state in a short time. Finally, the river water level, river flow and river wind speed along the river in the flood risk period are obtained.

[0076] Step S32: obtaining the historical river water level fluctuation and climate pattern, and performing water level evolution analysis on the river water level based on the historical river water level fluctuation and climate pattern to generate the corresponding river water level evolution fluctuation field in the flood risk period;

[0077] In the embodiment of the present application, by acquiring historical river water level fluctuation data and climate patterns and taking them as the basis for analysis, the evolution of the river water level is analyzed through historical observation data and climate simulation data. First, river water level fluctuation data for several years in the past are collected and analyzed. These data can be obtained from meteorological bureaus, hydrological departments and local weather stations. Then, future climate change trends, especially changes in precipitation and temperature, are predicted using a climate model. Combined with these historical data and climate patterns, a water level evolution analysis model is established. This model is based on the fluctuation rules of historical water level data to calculate the water level change trend during the flood risk period. In actual operation, time series analysis methods, machine learning regression models and other tools can be used to predict water level evolution and generate water level fluctuation fields during the period. Finally, the corresponding river water level evolution fluctuation field during the flood risk period is generated.

[0078] Step S33: The corresponding upstream inflow, midstream precipitation and downstream outflow are obtained by acquiring the river flow along the river during the flood risk period, and the flow correlation coupling analysis of the corresponding river water level change in the river water level evolution fluctuation field is carried out based on the upstream inflow, midstream precipitation and downstream outflow, to generate a dynamic response correlation field between the river flow and the water level change;

[0079] In the embodiment of the present application, by real-time monitoring of the river flow during the flood risk period, the flow data are obtained and further analyzed for the correlation between the flow and the upstream inflow, midstream precipitation and downstream outflow. First, a plurality of flow monitoring points are set up, and flow meters and flow meters and other equipment are used to obtain real-time flow data of each part of the river. Second, remote sensing monitoring and hydro-meteorological station data are used to estimate the upstream inflow and midstream precipitation, and the flow and water level evolution coupling analysis is carried out by inputting these data into the calculation model using the water balance equation in hydrology. By establishing the response function between flow change and water level fluctuation, the dynamic response correlation field is generated by combining the historical evolution trend of the water level, thereby providing a quantitative basis for subsequent flood risk warning. This process can use hydrological and hydraulic simulation software such as HEC-RAS to accurately analyze the correlation between flow and water level. Finally, the dynamic response correlation field between the river flow and the water level change along the river is generated.

[0080] Step S34: Based on the river flow along the river during the flood risk period, the wind speed influence factor evaluation analysis of the dynamic response correlation field between the river flow and the water level change is carried out, and the corresponding microscopic change influence factor of the river wind speed on the river water level and flow is obtained.

[0081] In the embodiment of the present application, by performing influence factor evaluation analysis based on wind speed data, first, actual wind speed data is collected along the river channel by wind speed monitoring instruments, wind speed sensors can be installed at different positions of the river channel, especially in areas susceptible to wind speed changes, combined with meteorological data, the micro influence of wind speed on river water level and flow changes is evaluated, the influence degree of wind speed on water surface and flow is determined through numerical simulation and fluid mechanics analysis, for example, wind speed can cause disturbance of water surface, thereby changing the distribution of water flow and water level change, using fluid dynamics model (such as CFD model), wind speed is added as an influence factor to the river water level and flow change model, thereby obtaining the quantitative influence factor of wind speed on water level and flow change, during the evaluation process, the micro change influence of different wind speed intervals on flow and water level can be quantified, and finally the micro change influence factor of wind speed along the river channel on river water level and flow is obtained.

[0082] Step S35: Based on the micro change influence factor of wind speed along the river channel on river water level and flow, flood risk prediction processing is performed on the corresponding river along the environmental water regime digital twin model, and river along the water regime flood risk prediction results are generated to perform corresponding river water regime flood risk management decision work.

[0083] In the embodiment of the present application, by using the micro change influence factor generated in the foregoing steps and the river water regime digital twin model to perform flood risk prediction during the flood risk period, the process first combines the water level, flow, wind speed and other data obtained in the previous steps, establishes a dynamic response model in the water regime digital twin model by fusing all influence factors, the model can predict water level and flow changes according to real-time monitoring data and historical trends, and adjust according to different flood situations, specifically, input real-time water level, flow and wind speed data, use numerical simulation method for prediction processing, output the prediction results of river water level and flow, use the prediction results to judge the flood risk level, and provide decision support for river management departments accordingly, the prediction system can generate specific flood risk warning report to guide water regime management and disaster prevention and mitigation work, the whole process can use computer simulation platform, such as MATLAB, Python programming language and special hydrological software, to perform comprehensive data processing and risk assessment, and finally perform corresponding river water regime flood risk management decision work.

[0084] Further, the flow correlation coupling analysis of the corresponding river along the water level change in the river along the water level evolution fluctuation field in step S33 based on the upstream inflow, the midstream precipitation and the downstream outflow includes the following steps:

[0085] Based on the upstream inflow and the midstream precipitation, gradient analysis of the upstream and midstream water flow along the river is carried out, and the distribution gradient of the upstream and midstream water flow along the river is obtained.

[0086] In the embodiment of the present application, the real-time data of the upstream inflow and the midstream precipitation are obtained, which are usually collected by meteorological satellites, basin hydrological stations and water conservancy monitoring systems. These data are the basis for the analysis of the river water flow. For the gradient analysis of the upstream water flow, the hydrodynamic model based on water level and flow is used for numerical simulation combined with the measured water flow data. In the specific operation, a plurality of water level monitoring points are set up on the upstream of the river. These monitoring points will update the water level change data in real time according to the actual flow and precipitation. Next, the influencing factors of the midstream water level change are calculated by the hydrodynamic model of the river, especially the contribution of the precipitation to the water level change. According to the model, the gradient analysis of the change of the water flow is carried out according to the flow velocity, flow, and the shape of the river (such as riverbed width, slope, etc.). The numerical solution method (such as finite difference method or finite element method) is used in the analysis process to gradually deduce the change distribution of the water flow between the upstream and the midstream. The change trend of the water flow between the upstream and the midstream of the river can be determined, and the evolution of the water flow can be reflected in real time. Finally, the distribution gradient of the upstream and midstream water flow along the river is obtained.

[0087] Preferably, based on the midstream precipitation and the downstream outflow, gradient analysis of the midstream and downstream water flow in the corresponding river along the water level evolution fluctuation field is carried out, and the distribution gradient of the midstream and downstream water flow along the river is obtained.

[0088] In the embodiment of the present application, the gradient analysis of the midstream and downstream water flow is carried out based on the midstream precipitation and the downstream outflow. First, the midstream precipitation and the downstream outflow data are obtained, which can be obtained through meteorological forecast, basin hydrological observation station and hydrological data platform of the reservoir management department. The midstream precipitation affects the change of the downstream water level, so it is necessary to analyze the direct influence of the precipitation on the water flow by the time series model. First, the midstream precipitation data are obtained in real time, and the influence of the midstream precipitation on the water level change is calculated combined with the hydrodynamic model of the water flow. Then, the downstream outflow data are obtained, which are monitored in real time by the flowmeter of the downstream river to ensure the accuracy of the data. The midstream precipitation and the downstream outflow are input into the model to calculate the change trend of the water flow of the midstream and downstream river section, and the gradient analysis of the water flow is carried out. In this process, the similar numerical method (such as finite element method) is used to analyze the gradient change of the midstream and downstream water flow, and the gradient distribution of the midstream and downstream water flow is obtained. Finally, the distribution gradient of the midstream and downstream water flow along the river is obtained.

[0089] Preferably, the upstream inflow and the downstream outflow are used to obtain the corresponding upstream and downstream flow error loss along the river channel, and the upstream and downstream flow error loss is used to analyze the water flow distribution loss of the upstream and downstream water flow change distribution gradient along the river channel, so as to generate the upstream-midstream-downstream water flow distribution gradient loss field along the river channel.

[0090] In the embodiment of the present application, the distribution loss of the water flow of each section of the river channel is analyzed by the upstream and downstream flow error loss. In actual operation, the difference between the upstream inflow and the downstream outflow often causes flow error, especially in the case of sudden rainfall or untimely water quantity scheduling. The flow error causes the violent fluctuation of water level change. Therefore, the upstream and downstream flow error is calculated first, that is, the water quantity difference between the upstream and downstream is analyzed by the flow balance formula. The calculation process is as follows: first, the upstream inflow, the downstream outflow and the water flow data of each section therebetween are obtained, the water quantity error between the upstream and downstream is calculated by the water flow model, the generation of the flow error can be estimated by simulating different water flow conditions, such as rainfall variation, river channel structure change, reservoir storage change and other factors. These errors will be optimized and corrected by specific mathematical models (such as least square method, Kalman filter, etc.), then the distribution loss of each section of water flow is calculated by using these errors, the flow loss and gain between different sections (upstream, midstream and downstream) of the river channel are considered, the loss field is obtained by the sectional analysis method, the water flow change loss of each section of the river channel is obtained by analyzing the error loss of the upstream inflow and the downstream outflow, so as to generate a detailed water flow distribution gradient loss field, and finally the upstream-midstream-downstream water flow distribution gradient loss field along the river channel is generated.

[0091] Preferably, the water level change amplitude analysis is performed on the corresponding water level change along the river channel in the water level evolution fluctuation field along the river channel, so as to obtain the water level change fluctuation amplitude along the river channel.

[0092] In the embodiment of the present application, by analyzing the water level change of each section of the river based on real-time water level monitoring data, the analysis process involves the amplitude, frequency and oscillation characteristics of water level fluctuation, and in specific operation, water level monitoring points need to be set up at multiple key nodes along the river, water level data is collected regularly, and water level change analysis is carried out in combination with weather forecast data. Next, the frequency distribution and amplitude of water level fluctuation are obtained by using frequency domain analysis method (such as fast Fourier transform FFT) to analyze the spectrum of water level change data. This analysis can reveal the strength of water level fluctuation and identify areas with large water level amplitude. Further, the amplitude and trend of water level change are determined through time domain analysis. In this process, other factors affecting water level change, such as river morphology, reservoir scheduling, downstream drainage, etc. need to be considered. By comprehensively considering these factors, the amplitude of water level change is analyzed in detail to ensure accurate prediction of water level change amplitude. Finally, the water level change fluctuation amplitude along the river is obtained.

[0093] Preferably, the water level change fluctuation amplitude along the river is analyzed by coupling the flow and water level change based on the upper-middle-lower stream water flow distribution gradient loss field along the river, to generate a dynamic response correlation field between the flow and water level change along the river.

[0094] In the embodiment of the present application, the coupling analysis between flow and water level change is carried out based on the upper-middle-lower stream water flow distribution gradient loss field and water level change fluctuation amplitude calculated in the previous step. To achieve this analysis, the correlation between water flow and water level change needs to be modeled, which can be done using multivariate regression analysis, neural network model, etc. to establish the dynamic response relationship between water flow and water level change. This analysis process will use correlation analysis method to calculate the relationship strength between flow change and water level change based on real-time water flow data and water level change data. On this basis, a dynamic response correlation field is established, which can reflect the influence of water flow change on water level change in real time and make a prediction of future water level change trend. Through this coupling analysis, the dynamic response correlation field obtained can provide accurate data support for river water regime prediction, help decision-makers take timely measures to cope with water level fluctuation, ensure the rational allocation and utilization of water resources, and finally generate a dynamic response correlation field between the flow and water level change along the river.

[0095] Further, step S34 includes the following steps:

[0096] Step S341: Based on the wind speed along the river during the flood risk period, the spatial distribution influence of the dynamic response correlation field between the flow and water level change along the river is analyzed, and the spatial distribution influence characteristics of the wind speed along the river on the water level and flow change of the river are obtained.

[0097] In the embodiment of the present application, by acquiring wind speed data along the river course, using high-precision weather models or weather data collection equipment such as automatic weather stations or satellite remote sensing, wind speed data of a specific period and location are obtained, which need to cover different positions of the river course, especially the wind speed changes of the upstream, midstream and downstream of the river course, ensure the spatial distribution representativeness of the data, match the wind speed data with the flow and water level change data of the river course, simulate and calculate the dynamic response of the river course flow and water level by hydrological and hydraulic models (such as HEC-RAS, MIKE11, etc.), obtain the influence relationship of wind speed on water level and flow change, and through dynamic correlation analysis, it can be identified that during the flood period, how the wind speed affects the water level and flow change of different positions along the river course, and by using spatial interpolation methods (such as Kriging interpolation or inverse distance weighted method), the spatial distribution influence atlas of wind speed on the water level and flow change of the river course is generated, so as to determine the specific influence characteristics of wind speed on water level and flow at different positions along the river course, and finally obtain the spatial distribution influence characteristics of wind speed on water level and flow change along the river course.

[0098] Step S342: acquiring the corresponding cross-section shape and slope of different river sections along the river course, and based on the cross-section shape and slope of different river sections along the river course, the spatial distribution influence degree of wind speed on water level and flow change between different river sections along the river course is calculated.

[0099] In the embodiment of the present application, by acquiring the river course data, the cross-section shape and slope of different river sections are further analyzed, which are usually obtained by topographic survey, remote sensing image or river survey, specifically, laser radar (LiDAR) technology or unmanned aerial photogrammetry can be used to accurately measure the cross-section shape and slope change of the river course, and by combining geographic information system (GIS) platform, the spatial analysis of river section cross-section shape and slope data is carried out to accurately describe the water flow behavior of different river sections and the spatial difference of wind speed influence, based on these river section shape characteristics, the influence of wind speed on water level and flow change is further evaluated by using hydrodynamic model (such as two-dimensional flow model or one-dimensional flow model), through model calculation, the influence degree of wind speed on water level and flow of different river sections can be obtained, and the difference of water level and flow response of each river section under wind speed change is revealed, a quantitative wind speed influence evaluation result is obtained, and finally the spatial distribution influence degree of wind speed on water level and flow change between different river sections along the river course is obtained.

[0100] Step S343: Based on the spatial distribution influence degree of the wind speed of different river sections along the river on the changes of the water level and flow of the river, the corresponding water level along the river in the dynamic response correlation field between the flow and water level changes along the river is evaluated and analyzed, and the corresponding micro change influence factor of the wind speed along the river on the water level and flow of the river is obtained.

[0101] In the embodiment of the present application, after obtaining the spatial distribution influence degree of the wind speed of different river sections on the changes of the water level and flow, further analysis is needed on the water level and flow data along the river to stratify different sections along the river according to the wind speed influence degree according to the spatial distribution influence degree. On this basis, the micro influence of wind speed on the changes of water level and flow is evaluated by calculating the wind speed influence factor. The specific implementation method is to analyze the correlation between wind speed data and water level and flow data, and to use regression analysis or machine learning algorithms (such as decision tree, support vector machine, etc.) to identify the influence factor of wind speed change on the changes of water level and flow. This process can be trained by historical data to establish the mapping relationship between wind speed and water level and flow, so as to obtain the specific influence coefficient of wind speed on the changes of water level and flow in different river sections and different time periods. In this process, the change of the influence factor needs to be evaluated according to different time scales (such as hour, day, month, etc.), so as to obtain the micro change influence factor of wind speed on water level and flow. Through this analysis, specific influence parameters can be accurately provided for dynamic water regime prediction in the digital twin model, and finally the corresponding micro change influence factor of wind speed along the river on the water level and flow of the river is obtained.

[0102] Further, step S4 includes the following steps:

[0103] Step S41: Obtain the water level along the river, the flow along the river and the temperature along the river corresponding to the dry risk period of the river environment water regime digital twin model in the dry risk estimation period of the river water regime.

[0104] In the embodiment of the present application, by using the water regime model based on digital twin technology to obtain the water level, flow and temperature data along the river course in the corresponding drought risk period range during the drought risk prediction period, a three-dimensional digital twin model of the river course is first constructed, including the geometric shape of the river course, historical hydrological data and real-time sensor data, combined with the meteorological prediction model and real-time observation data, the water level, flow and temperature data of the river course are collected in real time through remote sensing technology and Internet of Things sensors (such as water level sensor, flow meter, temperature sensor, etc.), then based on the digital twin platform, the predicted drought risk period (such as the next few days or weeks) and environmental change factors are input, the change trend of water level, flow and temperature in these periods is simulated, the future hydrological change is calculated through the existing historical data and climate model to ensure the accuracy of the output result, and finally the water level along the river course, the flow along the river course and the temperature along the river course in the drought risk period are obtained.

[0105] Step S42: Time series fluctuation cross and lag effect analysis is performed on the water level along the river course and the flow along the river course in the drought risk period to obtain the time series fluctuation cross interaction and time series fluctuation lag effect between the water level along the river course and the flow along the river course.

[0106] In the embodiment of the present application, after obtaining the water level and flow data along the river course, the time series fluctuation cross and lag effect are evaluated and analyzed, and the specific method is to use time series analysis technology to analyze the relationship between the river water level and flow in detail. First, the time series fluctuation cross effect between the water level and the flow is identified through statistical analysis methods such as correlation analysis and co-integration analysis, and the correlation between the water level and the flow at different time points is calculated through cross-correlation function (CCF) analysis to find out the interaction relationship between them. Secondly, the time lag of the influence of water level change on flow change is evaluated by using lag effect analysis, and the time lag effect of water level change on flow is quantified by using lag regression model or Granger causality test method. The lag effect analysis will help to identify whether there is a delay effect of flow responding to water level change in the drought risk period, and finally the time series fluctuation cross interaction and time series fluctuation lag effect between the water level along the river course and the flow along the river course are obtained.

[0107] Step S43: Based on the time series fluctuation cross interaction and time series fluctuation lag effect between the water level along the river course and the flow along the river course, time series fluctuation synchronization analysis is performed on the water level along the river course and the flow along the river course in the drought risk period to generate the corresponding river hydrological fluctuation response field in the drought risk period.

[0108] In the embodiment of the present application, after the time sequence fluctuation cross and hysteresis effect analysis of water level and flow is completed, the time sequence fluctuation synchronization analysis of water level and flow is carried out based on the analysis results, the synchronization analysis adopts synchronization analysis technology such as waveform correlation and phase synchronization analysis, and the synchronization and synchronization change degree of water level and flow change are quantified, and on this basis, the hydrological fluctuation response field along the river is generated through the simulation model, the fluctuation response mode between water level and flow in the drought risk period is evaluated, in the specific operation, the response field model is constructed by comparing the water level and flow data at different time periods and different positions, the hydrological change mode of the river in the drought risk period is identified, the generated hydrological fluctuation response field visually presents the change of hydrological conditions in the drought risk period, and finally the corresponding hydrological fluctuation response field along the river in the drought risk period is generated.

[0109] Step S44: Based on the temperature along the river, the hydrological evaporation rate of the corresponding hydrological fluctuation response field along the river in the drought risk period is evaluated and analyzed, and the influence correlation between the hydrological fluctuation along the river and the river surface evaporation rate is obtained.

[0110] In the embodiment of the present application, by evaluating the relationship between the temperature along the river and the hydrological fluctuation response field in the drought risk period, further analysis of the hydrological evaporation rate is carried out, first, based on the temperature data along the river, combined with the hydrological fluctuation response field, the evaporation calculation model (such as Penman-Monteith model or Hargreaves model) is used to evaluate the evaporation rate under different temperature conditions, the influence relationship between hydrological fluctuation and evaporation rate is quantified through the correlation between temperature, humidity, wind speed and other environmental factors and water level, flow fluctuation, for different river sections, according to the actual measured water surface temperature and environmental data, the corresponding evaporation rate is calculated, and through regression analysis, correlation analysis and other methods, the statistical model between water surface evaporation rate and water level fluctuation is obtained, this evaluation process has important guiding role for hydrological evaporation influence in drought period, can predict water loss and guide water resources management, and finally the influence correlation between the hydrological fluctuation along the river and the river surface evaporation rate is obtained.

[0111] Step S45: Based on the influence correlation between the hydrological fluctuation along the river and the river surface evaporation rate, the drought risk prediction processing of the corresponding environmental water regime digital twin model along the river is carried out, and the drought risk prediction result of the water regime along the river is generated to execute the corresponding drought risk management decision of the river water regime.

[0112] In the embodiment of the present application, by associating the influence between the previously obtained hydrological fluctuation and the evaporation rate, combining historical hydrological data and real-time monitoring data, the water regime along the river is drought risk forecasted, and the drought risk of the water regime is simulated through the digital twin model. Based on the obtained environmental data (such as water level, flow, temperature, etc.), real-time analysis and prediction are carried out, drought risk assessment is carried out using the model, the model predicts the future changes of river water level and flow according to the hydrological fluctuation response field and the water surface evaporation rate, simulates the range, intensity and duration of drought risk that may be affected, and outputs the risk prediction result. The result provides decision support for river water regime management. Especially when the drought risk is high, decision makers can arrange water resources dispatching, drought relief measures, etc. according to the prediction result, take relevant water regime management and emergency measures to ensure the rational use and protection of river water resources, and finally execute the corresponding river water regime drought risk management decision work.

[0113] Further, the present application also provides a river water regime prediction system based on digital twinning, which is used to execute the river water regime prediction method based on digital twinning as described above, and the river water regime prediction system based on digital twinning comprises:

[0114] The river water regime digital twinning modeling module is used to arrange sensor nodes along the river, and based on the sensor nodes, real-time monitoring of the river environment water regime along the river is carried out to obtain real-time data of the water level along the river, real-time data of the flow along the river, and real-time data of the weather along the river, wherein the real-time data of the weather along the river includes rainfall along the river, temperature along the river, and wind speed along the river; the real-time data of the water level along the river, the real-time data of the flow along the river, and the real-time data of the weather along the river are used to construct a digital twin along the river to generate an environmental water regime digital twin model along the river;

[0115] The river water regime risk estimation period division module is used to divide the corresponding river water regime period in the river water regime digital twinning model along the river based on the rainfall along the river to generate a river water regime flood risk estimation period and a river water regime drought risk estimation period;

[0116] The river water regime flood risk prediction module is used to obtain the corresponding water level along the river, flow along the river, and wind speed along the river in the river water regime flood risk estimation period, and based on the water level along the river, flow along the river, and wind speed along the river, the corresponding river water regime digital twinning model along the river is subjected to flood risk prediction processing, thereby generating a river water regime flood risk prediction result to execute corresponding river water regime flood risk management decision work;

[0117] The river water regime drought risk forecasting module is used for acquiring corresponding water levels along the river, flow along the river and temperature along the river in a river water regime drought risk estimation period, and performing drought risk forecasting processing on the corresponding environmental water regime digital twin model along the river based on the water levels along the river, the flow along the river and the temperature along the river, so as to generate a river water regime drought risk forecasting result, thereby performing corresponding river water regime drought risk management decision work.

[0118] The above description is merely that of the preferred embodiments of the application, and modifications of these embodiments will occur to those skilled in the art. All such modifications that do not depart from the spirit of the application are intended to be within the scope of the claims. The language used in the specification should not be used to limit the scope of the claims.

Claims

1. A river water regime forecasting method based on digital twins, characterized in that: The following steps are involved: Step S1: by deploying sensor nodes along the river channel, and based on the sensor nodes, real-time monitoring of the river environment and water conditions along the river channel is performed to obtain real-time water level data, real-time flow data, and real-time meteorological data along the river channel, wherein the real-time meteorological data along the river channel includes rainfall along the river channel, temperature along the river channel, and wind speed along the river channel; using the real-time water level data, real-time flow data, and real-time meteorological data along the river channel, a digital twin of the river channel is constructed to generate a digital twin model of the environmental water conditions along the river channel; Step S2: Based on the rainfall along the river, risk estimation is performed on the corresponding river water period in the digital twin model of the environmental water regime along the river to generate a river water flood risk estimation period and a river water drought risk estimation period; Step S3: Obtain the corresponding river water level, river flow, and river wind speed during the river water flood risk assessment period, and perform flood risk forecasting on the corresponding river water environment digital twin model based on the river water level, river flow, and river wind speed to generate a river water flood risk forecast result to implement corresponding river water flood risk management decision-making work; Step S4: Obtain the corresponding water level, flow rate and temperature along the river during the river water drought risk estimation period, and perform drought risk forecast processing on the corresponding river environmental water condition digital twin model based on the water level, flow rate and temperature along the river to generate a river water drought risk forecast result to execute corresponding river water drought risk management decision-making work.

2. The river water regime forecasting method based on digital twin according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: deploying sensor nodes along the river, wherein the sensor nodes include water level sensors, flow sensors, and meteorological monitoring sensors; Step S12: Based on the sensor nodes, the Internet of Things technology is used to connect the monitoring network to the cloud platform to generate an Internet of Things network for monitoring water conditions along the river; the water level sensors in the Internet of Things network for monitoring water conditions along the river are used to monitor the water level along the river in real time to obtain real-time water level data along the river; Step S13: using the flow sensors in the water regime monitoring IoT network along the river to monitor the flow along the river in real time, so as to obtain real-time data on the flow along the river; Step S14: using meteorological monitoring sensors within the river water regime monitoring IoT network to monitor the river's weather in real time, thereby obtaining real-time meteorological data along the river, including rainfall, temperature, and wind speed along the river; Step S15: Upload the real-time water level data, real-time flow data and real-time meteorological data along the river to the cloud platform, and use the real-time water level data, real-time flow data and real-time meteorological data along the river to construct a digital twin of the river to generate a digital twin model of the environmental water conditions along the river.

3. The river water regime forecasting method based on digital twin according to claim 2 is characterized in that: The construction of a digital twin of the river using the real-time water level data, flow data, and meteorological data along the river in step S15 includes the following steps: Conduct spatiotemporal correlation analysis on real-time water level data and flow data along the river to obtain the spatiotemporal dynamic correlation between water level changes and flow fluctuations along the river; Obtain the corresponding geographic information coordinate data along the river channel, and based on the spatiotemporal dynamic correlation between water level changes and flow fluctuations along the river channel, perform spatiotemporal water regime correlation mapping on the real-time water level data and flow data along the river channel in combination with the geographic information coordinate data to generate a spatiotemporal water regime correlation feature model along the river channel; Based on the real-time meteorological data along the river, the spatiotemporal correlation characteristic model of the water regime along the river is analyzed by fusion of meteorological multi-source data to generate a time series model of the environmental water regime along the river; Obtain digital terrain data corresponding to the river channel, including riverbed slope and shoreline contours, and perform river basin terrain simulation analysis on the digital terrain data corresponding to the river channel to generate river basin terrain and water flow simulation data along the river channel; Based on the terrain and water flow simulation data of the river basin along the river, a digital twin fusion of the environmental water condition time series model along the river is constructed to generate a digital twin model of the environmental water condition along the river.

4. The river water regime forecasting method based on digital twin according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: Based on the rainfall along the river, a spatiotemporal fluctuation correlation analysis of the water regime of the corresponding river water regime period in the digital twin model of the environmental water regime along the river is performed to obtain a spatiotemporal fluctuation correlation relationship between the rainfall along the river and the water regime fluctuation; Step S22: Based on the spatiotemporal correlation between rainfall and water regime fluctuations along the river channel, a water regime fluctuation time series feature analysis is performed on the rainfall along the river channel and the corresponding river water regime period in the digital twin model of the environmental water regime along the river channel, thereby obtaining a rainfall-water regime fluctuation time series feature dataset along the river channel, which includes periodic and non-periodic fluctuation features corresponding to rainfall, flow fluctuations, and water storage changes. Step S23: Based on the rainfall-water regime fluctuation time series feature dataset along the river channel, a water regime fluctuation risk determination calculation formula is used to perform risk determination and assessment calculation for the corresponding river water regime period in the environmental water regime digital twin model along the river channel, so as to obtain water regime fluctuation risk determination parameters along the river channel, including flood fluctuation risk determination parameters along the river channel and drought fluctuation risk determination parameters along the river channel; Step S24: Based on the water regime fluctuation risk judgment parameters along the river channel, the corresponding river water regime periods in the digital twin model of the environmental water regime along the river channel are divided into flood and drought extreme risks. If the river water regime fluctuation in the corresponding river water regime period is within the range corresponding to the flood fluctuation risk judgment parameters along the river channel, then the river water regime period is determined as the river water regime flood risk estimation period; if the river water regime fluctuation in the corresponding river water regime period is within the range corresponding to the drought fluctuation risk judgment parameters along the river channel, then the river water regime period is determined as the river water regime drought risk estimation period.

5. The river water regime forecasting method based on digital twin according to claim 4 is characterized in that: The calculation formula for determining the water regime fluctuation risk in step S23 is specifically: Where R H is the flood fluctuation risk determination parameter along the river, R G is the drought fluctuation risk determination parameter along the river, t0 is the initial time point corresponding to the river water period, t1 is the end time point corresponding to the river water period, t is the time variable parameter, P(t) is the rainfall at time t, is the rainfall change rate, α1 is the rainfall water regime fluctuation risk weight coefficient, Q(t) is the river flow at time t, Q max is the maximum threshold of river flow, H(t) is the change of river water storage at time t, H max is the maximum threshold of river water storage change, α2 is the risk weight coefficient of water storage change and water regime fluctuation, exp is the exponential function, β is the risk weight coefficient of river flow and water regime fluctuation, n is the total number of high tide events, i is the item index of high tide events, Q i is the river discharge of the ith high tide event, Q avg is the average river flow of high tide events, m is the total number of drought events, j is the index of drought events, S j is the standardized drought evaporation index of the jth drought event, S avg is the average evaporation index of drought events, η1 is the correction coefficient of the flood fluctuation risk determination parameter along the river channel, and η2 is the correction coefficient of the drought fluctuation risk determination parameter along the river channel.

6. The river water regime forecasting method based on digital twin according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: obtaining the water level, flow rate and wind speed along the river during the flood risk period corresponding to the digital twin model of the environmental water conditions along the river during the river water condition flood risk estimation period; Step S32: Obtain historical river water level fluctuations and climate patterns, and perform water level evolution analysis on the water level along the river based on the historical river water level fluctuations and climate patterns to generate a corresponding water level evolution fluctuation field along the river during the flood risk period; Step S33: Obtaining the corresponding upstream water inflow, midstream precipitation, and downstream water outflow based on the flow along the river during the flood risk period, and performing flow correlation coupling analysis on the corresponding water level changes along the river in the water level evolution fluctuation field along the river based on the upstream water inflow, midstream precipitation, and downstream water outflow, so as to generate a dynamic response correlation field between the flow along the river and the water level changes; Step S34: performing wind speed impact factor evaluation and analysis on the dynamic response correlation field between flow and water level changes along the river channel based on the wind speed along the river channel during the flood risk period, and obtaining the microscopic impact factor of the wind speed along the river channel on the river channel water level and flow; Step S35: Based on the micro-change influencing factors of the water level and flow along the river channel, the flow along the river channel, and the wind speed along the river channel, the corresponding digital twin model of the environmental water conditions along the river channel is processed for flood risk forecasting, and the water condition flood risk forecast results along the river channel are generated to execute the corresponding river water condition flood risk management decision-making work.

7. The river water regime forecasting method based on digital twin according to claim 6 is characterized in that: The flow correlation coupling analysis of the corresponding water level changes along the river channel in the water level evolution fluctuation field along the river channel based on the upstream water inflow, midstream precipitation and downstream water outflow in step S33 includes the following steps: Based on the upstream water volume and midstream precipitation, the upper and middle water flow gradients in the middle reaches of the river corresponding to the water level evolution fluctuation field along the river are analyzed to obtain the distribution gradient of water flow changes in the middle reaches of the river. Based on the midstream precipitation and downstream water discharge, the midstream and downstream water flow gradient analysis is carried out along the corresponding midstream and downstream of the river in the water level evolution fluctuation field along the river channel, and the distribution gradient of the water flow change along the midstream and downstream of the river channel is obtained; The upstream inflow and downstream outflow are used to obtain the corresponding upstream and downstream flow error losses along the river channel. Based on the upstream and downstream flow error losses along the river channel, the water flow distribution loss of each section of the river channel is analyzed for the water flow change distribution gradient in the middle reaches of the river channel and the water flow change distribution gradient in the middle and lower reaches of the river channel to generate the water flow distribution gradient loss field along the upper, middle and lower reaches of the river channel. The water level change amplitude analysis is performed on the water level change along the river channel corresponding to the water level evolution fluctuation field along the river channel to obtain the water level change fluctuation amplitude along the river channel; Based on the gradient loss field of water flow distribution in the upper, middle and lower reaches of the river, a flow correlation coupling analysis of the fluctuation amplitude of water level changes along the river is carried out to generate a dynamic response correlation field between flow and water level changes along the river.

8. The river water regime forecasting method based on digital twin according to claim 6 is characterized in that: Step S34 includes the following steps: Step S341: performing a spatial distribution impact analysis based on the dynamic response correlation field between the wind speed along the river channel and the flow and water level changes along the river channel during the flood risk period, and obtaining the spatial distribution impact characteristics of the wind speed along the river channel on the river channel water level and flow changes; Step S342: Obtaining the cross-sectional morphology and slope of different river sections along the river channel, and performing river section wind speed impact assessment calculation based on the spatial distribution of wind speed impact on river water level and flow changes along the river channel based on the cross-sectional morphology and slope of different river sections along the river channel, to obtain the spatial distribution of wind speed impact on river water level and flow changes along the river channel; Step S343: Based on the spatial distribution of the impact of wind speed on the changes in river water level and flow along different river sections, the wind speed impact factor is evaluated and analyzed on the corresponding water level along the river and the flow along the river in the dynamic response correlation field between the flow along the river and the water level change, so as to obtain the microscopic change impact factor of the wind speed along the river on the river water level and flow.

9. The river water regime forecasting method based on digital twin according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: obtaining the water level, flow rate, and temperature along the river during the drought risk period corresponding to the digital twin model of the environmental water conditions along the river during the river water drought risk estimation period; Step S42: performing a time series fluctuation cross and hysteresis evaluation analysis on the water level and flow along the river during the drought risk period, and obtaining the time series fluctuation cross interaction and time series fluctuation hysteresis effect between the water level and flow along the river; Step S43: Based on the cross-interaction of the temporal fluctuations between the water level and flow along the river channel and the hysteresis effect of the temporal fluctuations, the temporal fluctuations of the water level and flow along the river channel during the drought risk period are synchronously analyzed to generate a hydrological fluctuation response field along the river channel corresponding to the drought risk period; Step S44: performing a hydrological evaporation rate assessment and analysis on the hydrological fluctuation response field along the river corresponding to the drought risk period based on the temperature along the river, and obtaining an impact correlation relationship between the hydrological fluctuation along the river and the evaporation rate of the river surface; Step S45: Based on the influence correlation between the hydrological fluctuations along the river and the evaporation rate of the river surface, the corresponding river water environment digital twin model is processed for drought risk forecasting, and the water condition drought risk forecast results along the river are generated to execute the corresponding river water condition drought risk management decision-making work.

10. A river water regime forecasting system based on digital twins, characterized in that: For executing the river water regime forecasting method based on digital twin according to claim 1, the river water regime forecasting system based on digital twin comprises: The river water regime digital twin modeling module is used to deploy sensor nodes along the river channel and conduct real-time monitoring of the river environment and water regime along the river channel based on the sensor nodes to obtain real-time data on water level, flow, and meteorology along the river channel, including rainfall, temperature, and wind speed along the river channel. The module uses the real-time water level, flow, and meteorology data along the river channel to construct a digital twin of the river channel to generate a digital twin model of the river environment and water regime. The river water regime risk estimation period division module is used to perform risk estimation and division of the corresponding river water regime periods in the digital twin model of the environmental water regime along the river based on the rainfall along the river, so as to generate the river water regime flood risk estimation period and the river water regime drought risk estimation period; The river water regime and flood risk forecasting module is used to obtain the corresponding river water level, river flow and river wind speed during the river water regime and flood risk assessment period, and perform flood risk forecasting on the corresponding river environmental water regime digital twin model based on the river water level, river flow and river wind speed, thereby generating the river water regime and flood risk forecast results to implement the corresponding river water regime and flood risk management decision-making work; The river water condition and drought risk forecast module is used to obtain the corresponding water level, flow rate and temperature along the river during the river water condition and drought risk estimation period, and perform drought risk forecast processing on the corresponding river environmental water condition digital twin model based on the water level, flow rate and temperature along the river, thereby generating the river water condition and drought risk forecast results to execute the corresponding river water condition and drought risk management decision-making work.

Images