A method and system for real-time water regime forecasting

Abstract

The present invention relates to the technical field of real-time water situation forecasting, and in particular to a method and system for real-time water situation forecasting. The method comprises the following steps: acquiring real-time hydrological data; performing real-time hydrological data analysis based on the real-time hydrological data to obtain real-time hydrological analysis data; constructing a water situation forecasting model based on the real-time hydrological analysis data to obtain a water situation forecasting model; generating hydrological warning data based on the water situation forecasting model to obtain hydrological warning data to perform real-time water situation forecasting operations. The present invention can not only capture hydrological changes in real time, but also dynamically respond to the influence of meteorological conditions and complex water flow paths, generate high-precision warning data, effectively help decision makers deal with flood risks, and significantly improve the accuracy, real-time and adaptability of water situation forecasting, especially when dealing with sudden floods or extreme weather events.

Description

Technical Field

[0001] The present invention relates to the technical field of real-time water regime forecasting, and in particular to a real-time water regime forecasting method and system. Background Art

[0002] Water forecasting systems play an important role in flood control, drainage, irrigation management, and urban water resources management. As climate change leads to frequent extreme weather events, such as heavy rains, floods, droughts and other disasters, accurate and real-time water forecasting has become an important tool for water management departments and related industries to deal with these risks. Traditional water forecasting systems rely on historical data and statistical models, and cannot respond to the dynamic changes of complex hydrological conditions in real time, nor can they adapt to sudden extreme weather events.

[0003] In the field of water forecasting, the accuracy of hydrological data acquisition, analysis and forecasting models is crucial to the practicality of the system. However, due to the complexity of the hydrological system, factors such as the linkage effect of multi-location data, changes in terrain and meteorological conditions, and the complexity of water flow paths, traditional hydrological analysis methods often cannot meet the requirements of real-time and accurate prediction. Especially under complex terrain conditions, the path and transmission characteristics of water flow are difficult to accurately simulate, resulting in low forecast accuracy. In addition, traditional methods have poor real-time performance, are difficult to cope with sudden hydrological changes, have delayed warnings, and lack the ability to respond quickly. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a real-time water situation forecasting method and system to solve at least one of the above technical problems.

[0005] The present application provides a method for real-time water situation forecasting, comprising the following steps:

[0006] Step S1: obtaining real-time hydrological data;

[0007] Step S2: performing real-time hydrological data analysis based on the real-time hydrological data to obtain real-time hydrological analysis data;

[0008] Step S3: constructing a water regime forecast model based on real-time hydrological analysis data to obtain a water regime forecast model;

[0009] Step S4: Generate hydrological warning data according to the water regime forecast model to obtain hydrological warning data for real-time water regime forecasting.

[0010] The real-time data acquisition in the present invention ensures the timeliness of the forecast, can quickly respond to sudden changes in water conditions, and reduce the risks caused by delays. Through hydrological data analysis, the trend of water level rise or flow rate change can be discovered in advance, and potential risks such as floods and droughts can be quickly predicted, thereby improving the predictability of sudden hydrological events. The model can continuously optimize the forecast accuracy and improve the response ability to hydrological dynamic changes through the combination of historical data and real-time data. It can adapt to different hydrological environments and climatic conditions to improve the accuracy and stability of forecasts. Through the data predicted by the water situation forecast model, the system can automatically generate hydrological early warning information. When the water level, flow or rainfall reaches a certain early warning threshold, the system can issue an early warning notification in time to prevent or reduce the impact of water disasters.

[0011] Preferably, step S1 specifically includes:

[0012] Collecting hydrological data through a hydrological monitoring device preset at a first geographical location to obtain first real-time hydrological data;

[0013] Collecting hydrological data through a hydrological monitoring device preset at a second geographical location to obtain second real-time hydrological data;

[0014] The first real-time hydrological data and the second real-time hydrological data are integrated to obtain real-time hydrological data.

[0015] The wide coverage in the present invention makes the hydrological data more comprehensive, reduces the local errors or blind spots that may be caused by single-point data collection, and can monitor the hydrological changes in different regions. Through the integration of data from multiple monitoring points, more stable and accurate hydrological data can be obtained, ensuring that the basic data for analysis and forecasting are more reliable, providing hydrological conditions in different locations, thereby improving the accuracy of the overall forecast. Through real-time monitoring and data integration, rapid changes in water conditions can be discovered in a timely manner, and the responsiveness of the system is enhanced, especially early warning of sudden hydrological events. Since the data comes from different geographical locations, the system can use these data to perform hydrological linkage analysis between regions. For example, changes in the upstream water level affect the water flow and water level downstream. Through the linkage of data from different locations, the hydrological changes of the entire water system can be better analyzed and predicted.

[0016] Preferably, step S2 specifically includes:

[0017] Constructing a hydrological correlation matrix according to the first real-time hydrological data and the second real-time hydrological data to obtain hydrological correlation matrix data;

[0018] Perform geographic location factor analysis based on hydrological correlation matrix data to obtain hydrological correlation geographic location analysis data;

[0019] Perform location linkage analysis based on hydrologically associated geographical location analysis data to obtain hydrologically associated geographical linkage analysis data;

[0020] Real-time hydrological analysis data is generated based on the hydrologically associated geographic linkage analysis data to obtain real-time hydrological analysis data.

[0021] In the present invention, by constructing a hydrological association matrix, the correlation of hydrological characteristics between different geographical locations can be quantified. The association matrix can reflect the correlation and mutual influence of hydrological data between two locations, especially in capturing the synchronization of hydrological changes in a dynamic environment. Through the construction of the association matrix, the accuracy of hydrological data analysis can be effectively improved. The mutual correlation of hydrological data at different locations enables the system to better understand the hydrological interaction between different regions, providing a solid data foundation for linkage analysis. Considering the geographical location factor can make the analysis results more in line with the actual situation and increase the accuracy and practicality of the forecast model. By combining geographical features, the system can identify and explain hydrological changes under different geographical environments, so as to better adapt to the complex natural environment. Location linkage analysis can provide a global perspective, so that the system not only pays attention to local hydrological conditions, but also comprehensively analyzes the linkage relationship between various locations in the region, better predicts the dynamic changes of hydrology in the entire basin or region, and provides more reliable forecast results. The generation of real-time hydrological analysis data ensures that the system can continuously adjust and update its analysis results, ensuring that the forecast model is still effective under dynamically changing hydrological conditions. The ability to update real-time data improves the timeliness of the system and can quickly respond to changing hydrological conditions.

[0022] Preferably, the hydrological correlation matrix is ​​constructed as follows:

[0023] Extracting hydrological characteristics according to the first real-time hydrological data and the second real-time hydrological data to obtain first real-time hydrological characteristic data and second real-time hydrological characteristic data;

[0024] Constructing a terrain undulation correlation matrix according to the first real-time hydrological characteristic data and the second real-time hydrological characteristic data to obtain real-time hydrological terrain undulation correlation matrix data;

[0025] Constructing a watershed distribution association matrix according to the first real-time hydrological characteristic data and the second real-time hydrological characteristic data to obtain real-time hydrological watershed distribution association matrix data;

[0026] The initial hydrological correlation matrix data is obtained by performing correlation calculation based on the real-time hydrological terrain fluctuation correlation matrix data and the real-time hydrological watershed distribution correlation matrix data;

[0027] The initial hydrological correlation matrix data is subjected to correlation matrix sparse processing to obtain the hydrological correlation matrix data.

[0028] In the present invention, by extracting key hydrological features (such as water level, flow, rainfall, flow rate, etc.) from the real-time hydrological data of the first and second geographical locations, the dynamic changes of the hydrological system can be captured, and real-time hydrological feature data for analysis can be generated. The terrain undulation analysis introduces the physical characteristics of the hydrological system, ensures that the impact of different terrains on hydrological changes is taken into account in the analysis process, improves the understanding of hydrological flow and transmission laws, and enhances the physical basis of the association matrix. Combined with the characteristics of watershed distribution, the distribution characteristics in the hydrological system and the linkage relationship between watersheds can be better explained, especially in the complex environment of large watersheds and multiple sub-watersheds, the watershed distribution association matrix can help understand the transmission, convergence and dispersion process of water flow. Multi-dimensional association analysis enhances the adaptability of the system to complex hydrological environments, enables it to identify the relationship between different hydrological features, and analyzes multiple physical conditions (such as terrain, watershed distribution) to make the association matrix more systematic and accurate. It effectively enhances the understanding of the mutual influence between hydrological features in multiple locations, especially in the interaction between upstream and downstream and different watersheds, which can reveal the complex hydrological linkage mechanism and improve the prediction accuracy of the system. Sparsification processing not only improves computational efficiency, especially when facing large-scale hydrological data, by removing elements in the correlation matrix that are not significantly correlated with hydrological changes and retaining those hydrological features with significant correlation, but also enhances the robustness of the model and the accuracy of the overall analysis by reducing unnecessary noise and redundant data.

[0029] Preferably, the geographical location factor analysis is specifically as follows:

[0030] Conduct terrain and topography impact analysis based on hydrological correlation matrix data to obtain terrain impact data;

[0031] Conduct soil and vegetation type analysis based on terrain impact data to obtain soil and vegetation impact data;

[0032] Meteorological factors are analyzed based on soil and vegetation impact data to obtain hydrologically related geographical location analysis data.

[0033] Considering the terrain and topographic factors in the present invention can ensure that the hydrological analysis is more in line with the actual situation, especially it can reveal the hydrological linkage characteristics between different geographical locations, for example, areas with large slopes lead to rapid water flow, while gentle terrain produces water retention. The analysis of soil and vegetation types further deepens the understanding of the hydrological system, so that the model can accurately reflect the treatment of water in different regions. For example, areas with strong soil permeability lead to rapid water infiltration, reducing surface runoff, while areas with more vegetation coverage will significantly slow down the surface runoff rate. The combination of meteorological factor analysis and geographical location factors enables the system to link the driving force of hydrological changes with specific environmental conditions, which helps to explain the hydrological changes under meteorological conditions. For example, in the case of heavy rain, some areas are more likely to accumulate water or produce runoff, thereby improving the accuracy of the forecast. The refinement of local hydrological forecasts can provide more targeted forecast results for different regions. Whether in mountainous areas, plains or lake areas, the differences in the impact of different geographical features on hydrology have been fully considered, so that the system can better adapt to various complex environments. The analysis method of global and local linkage can significantly improve the comprehensiveness of the hydrological forecast for the entire region. It can not only provide accurate hydrological forecasts for local areas, but also provide clearer analysis of the hydrological linkage within the entire river basin, thereby achieving more accurate global forecasts.

[0034] Preferably, the location linkage association is specifically:

[0035] Perform time-lag correlation analysis based on hydrologically associated geographical location analysis data to obtain hydrologically associated time-lag data;

[0036] The location transmission model is constructed based on the hydrological location time-lag correlation data to obtain the hydrological transmission model;

[0037] The linkage intensity is processed according to the hydrological transmission model to obtain the hydrologically associated geographic linkage analysis data.

[0038] In the present invention, by analyzing the time lag between hydrological data in different geographical locations, the propagation speed and hysteresis effect of hydrological events can be identified. For example, there is a time lag in water level changes between the upstream and downstream of a river, and the time lag effect reflects the transmission time of the water flow. By quantifying this lag time, the dynamic process of hydrological linkage can be understood more accurately. A hydrological transmission model is constructed based on the hydrological position time-lag correlation data. The model simulates the hydrological transmission process between different locations and can describe key parameters such as the transmission path, speed and residence time of the water flow. By combining topography, watershed characteristics and hydrological time series data, the transmission model provides an explanation of the physical mechanism of hydrological changes. The linkage intensity processing can quantify the hydrological interactions between different locations, help identify the impact of hydrological changes in key areas (such as upstream areas) on other areas, and not only help to improve the accuracy of hydrological forecasts, but also provide a more accurate basis for water resources management and scheduling. Cross-regional hydrological linkage analysis can improve the hydrological forecasting capabilities over a large area, especially in environments with complex hydrological changes between upstream and downstream and multiple river basins. The system can accurately predict the hydrological interactions between different regions and enhance the spatial breadth of the forecast.

[0039] Preferably, the linkage strength processing is specifically as follows:

[0040] The mixed linkage intensity is processed according to the hydrological transmission model to obtain the first hydrologically associated geographic linkage analysis data;

[0041] Generate multi-level hysteresis coefficients according to the hydrological transmission model to obtain hysteresis coefficient data;

[0042] The hydrological transmission model is subjected to time series regression calculation according to the lag period data to obtain the hydrological time series regression model;

[0043] The Bayesian network causal relationship processing is carried out according to the hydrological time series regression model to obtain the second hydrological associated geographic linkage analysis data;

[0044] The first hydrologically associated geographic linkage analysis data and the second hydrologically associated geographic linkage analysis data are adjusted based on similarities and differences to obtain hydrologically associated geographic linkage analysis data.

[0045] In the present invention, by combining different linkage analysis methods (such as linear and nonlinear methods), the system can more comprehensively capture the linkage strength of hydrological data between different locations. Hybrid linkage strength processing can simultaneously analyze simple linear relationships and complex nonlinear associations, ensuring that the system can accurately identify the linkage between locations under a variety of hydrological conditions. Through multi-level lag numbers, the system can capture the dynamically changing spatiotemporal correlation characteristics, making the system more adaptable to the delay effect in the hydrological system, which plays a key role in accurately predicting the water flow transmission time, waterlogging risk, etc., and improves the timeliness and accuracy of the forecast model. The time series regression model can provide long-term and short-term trend forecasts for hydrological forecasts, especially with the support of multi-level lag data, it can more accurately predict the hydrological conditions at a certain point in the future. Bayesian network causal analysis helps to reveal the causal mechanism behind hydrological linkage, especially in complex watershed systems, and can help identify which locations are key nodes or areas that have an important impact on downstream. The difference feedback adjustment mechanism enables the system to dynamically optimize its own analysis results, enhancing the accuracy and stability of the forecast results. By combining the advantages of multiple analysis methods and automatically adjusting inconsistent results, the system can reduce errors and improve its adaptability to complex hydrological environments.

[0046] Preferably, the real-time hydrological analysis data generation is specifically as follows:

[0047] Decompose the time series characteristics according to the hydrologically associated geographic linkage analysis data to obtain the hydrologically associated time series characteristics decomposition data;

[0048] According to the hydrological correlation time series feature decomposition data, dynamic spatial mapping is performed to obtain hydrological correlation spatial reconstruction data;

[0049] Embedding local meteorological factors according to the hydrological correlation spatial reconstruction data to obtain local meteorological factor embedded data;

[0050] The water flow path model is constructed according to the local meteorological factor embedded data to obtain the water flow path model;

[0051] Multi-point collaborative processing is performed based on the water flow path model to obtain real-time hydrological analysis data.

[0052] The decomposition of time series features in the present invention provides a multi-level understanding for hydrological analysis, which can more accurately predict long-term trends and short-term fluctuations in the hydrological system, and improve the ability to control complex hydrological changes. Dynamic spatial mapping improves the spatial understanding of hydrological data, and can identify the expansion path of water flow in different geographical areas and its impact on different locations. It provides an accurate spatial background for subsequent hydrological analysis and helps to optimize the results of spatial analysis. By embedding meteorological factors, the system can dynamically respond to weather changes and can better predict the changes in hydrological characteristics under different meteorological conditions, which helps to improve the hydrological response capabilities under extreme weather, especially the early warning of rainstorms, droughts and other situations. According to the meteorological factor embedded data, the water flow path model can simulate the flow path, speed and retention area of ​​water under different terrain and meteorological conditions, which can not only describe the physical flow process of water, but also accurately predict the dynamic changes of different locations where water flows. Multi-point collaborative processing enables the system to have stronger regional linkage analysis capabilities, and can predict the linkage of hydrological changes between upstream and downstream and different river basins, providing a more comprehensive perspective for cross-regional water resources management and risk prediction, and helping to improve the spatial breadth and accuracy of the overall forecast.

[0053] Preferably, step S3 specifically includes:

[0054] Conduct finite element analysis based on real-time hydrological analysis data to obtain real-time hydrological finite element data;

[0055] Perform hydrodynamic calculations based on real-time hydrological finite element data to obtain real-time hydrological and hydrodynamic data;

[0056] The water regime forecast model is constructed according to the real-time hydrological and hydrodynamic data to obtain the water regime forecast model.

[0057] In the present invention, the finite element analysis method is used to process real-time hydrological data, and the complex hydrological spatial distribution can be discretized, and the study area can be divided into multiple small units (i.e., finite elements), each of which can reflect the changes in hydrological characteristics in a specific area. According to the real-time hydrological finite data, hydrodynamic calculations can be performed to accurately simulate the flow characteristics of water bodies at different time and space scales, such as water flow velocity, flow, water level changes, etc. Hydrodynamic calculations can handle a variety of complex water flow forms, such as turbulence in rivers and overflow on plains, and are suitable for simulating the hydrodynamic effects of floods or extreme rainfall events. Model construction based on hydrodynamic data can dynamically transform complex hydrological changes into predictable patterns and generate a water situation forecast model with high precision. The model combines the results of spatial finite element analysis and hydrodynamic calculations, and can perform real-time and accurate water situation forecasts to adapt to hydrological changes at different time and space scales. The terrain and hydrological characteristics are highly integrated through finite element analysis, and the water flow path is simulated through hydrodynamic calculations, which is suitable for hydrological systems under complex terrain conditions, such as mountainous areas, plains, rivers, lakes, etc. The system can effectively predict hydrological changes under different terrain conditions, improving the environmental adaptability of the model.

[0058] Preferably, the present application also provides a water condition real-time forecasting system for executing the water condition real-time forecasting method as described above, the water condition real-time forecasting system comprising:

[0059] Real-time hydrological data acquisition module, used to obtain real-time hydrological data;

[0060] A real-time hydrological data analysis module is used to perform real-time hydrological data analysis based on real-time hydrological data to obtain real-time hydrological analysis data;

[0061] A water regime forecast model construction module is used to construct a water regime forecast model based on real-time hydrological analysis data to obtain a water regime forecast model;

[0062] The hydrological warning data generation module is used to generate hydrological warning data according to the water situation forecast model to obtain hydrological warning data for real-time water situation forecasting operations.

[0063] The beneficial effect of the present invention is that the entire process from real-time hydrological data acquisition to early warning data generation is automated, reducing human intervention, ensuring that emergencies can be captured in time and responded quickly during hydrological changes. Through real-time monitoring, analysis and forecasting, the system can instantly generate hydrological early warning information and automatically transmit it to relevant departments so that emergency measures can be taken. Multi-level analysis can adapt to complex hydrological environments, such as hydrological changes under various different geomorphic conditions such as mountainous areas, plains, and lakes. When dealing with complex natural environments, the system can not only process local hydrological data, but also comprehensively analyze the hydrological linkage and geographical factors of multiple locations, provide a deep insight combining global and local, and improve the accuracy of hydrological forecasts. By introducing spatial and dynamic characteristics into the forecast model, the system can simulate the dynamic changes of water flow under different geographical conditions, and accurately predict the transmission path, speed and water level changes of water flow, adapt to various complex terrain conditions, such as mountain rivers, urban drainage systems, etc., and improve the environmental adaptability and prediction ability of the forecast model. By integrating Bayesian network causal relationship analysis into the hydrological forecast process, the system can identify the causal relationship between hydrological changes based on historical data and real-time data. For example, a sharp change in rainfall upstream causes a rise in water levels downstream. The system can predict potential flood risks in advance based on this causal relationship. Through multi-point collaborative processing and cross-regional hydrological data analysis, the hydrological linkage between different locations can be identified, especially the flow transmission and water level changes in large river basins and upstream and downstream water systems, which can provide a scientific basis for cross-regional water resources management and water regime scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting implementations made with reference to the following drawings:

[0065] Figure 1 A flow chart showing the steps of a method for real-time water situation forecasting according to an embodiment is shown;

[0066] Figure 2 A flowchart showing a method for real-time hydrological data collection according to an embodiment is shown;

[0067] Figure 3 A flowchart showing a method for real-time hydrological data analysis according to an embodiment is shown;

[0068] Figure 4 A flow chart showing the steps of a method for generating hydrological warning data according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0069] The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0070] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0071] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0072] The system obtains hydrological data in real time through preset hydrological monitoring equipment, such as water level, flow rate, rainfall, evaporation, etc. These data come from different geographical locations (such as upstream and downstream monitoring stations). The monitored data are: upstream water level: 2.3 meters, downstream water level: 1.5 meters, upstream flow rate: 1.2 meters / second, downstream flow rate: 0.8 meters / second, rainfall: 10 mm / hour (evenly distributed).

[0073] The collected hydrological data were analyzed in time series to separate long-term trends, cyclical changes and short-term fluctuations. After analyzing the upstream water level data, it was found that the long-term trend was a slow rise in water level (0.05 meters per hour), and the short-term fluctuation was a rapid change in water level caused by rainfall (10 mm of rainfall increased the water level by 0.1 meters). The upstream and downstream data were linked and analyzed, and spatial mapping was constructed through the Geographic Information System (GIS) to understand the flow path and time lag of water flow in the upstream and downstream. The analysis showed that the time lag for water transmission from upstream to downstream was 2 hours, and the flow path length was 10 kilometers. Meteorological data such as rainfall and evaporation were embedded in the analysis to evaluate the impact of meteorological factors on hydrological changes. The current rainfall is 10 mm / hour, and the rainfall intensity is expected to increase to 30 mm / hour in the next 3 hours, increasing the risk of flooding.

[0074] The water area is discretized and the basin is divided into multiple small units to analyze the spatial changes of water level, flow velocity and other characteristics. The basin is divided into 100 small units, each representing an area of ​​100 meters × 100 meters. Finite element analysis shows that in the next 2 hours, the upstream water level will increase to 2.8 meters and the downstream water level will increase to 2.0 meters. Based on finite element data, hydrodynamic calculations are performed to simulate the flow characteristics of the water body and predict water flow velocity, water level changes, etc. Hydrodynamic calculations show that the upstream water flow velocity will increase to 1.5 meters / second, the downstream water flow velocity will increase to 1.1 meters / second, the flood peak will reach the downstream in 3 hours, and the downstream water level is expected to rise to 2.3 meters. A water situation forecast model is constructed based on hydrodynamic data to predict water situation changes in the next few hours. The forecast model shows that rainfall will continue in the next 6 hours, and the upstream water level is expected to rise to 3.1 meters and the downstream water level will reach 2.7 meters, which poses a flood risk.

[0075] According to the water situation forecast model, the system generates hydrological warning data, sets the corresponding warning water level threshold, and generates warning notifications. According to the forecast results, the system sets the warning threshold to 2.5 meters. When the downstream water level reaches or exceeds this value, the system will issue a flood warning.

[0076] There are two monitoring stations in the basin, upstream and downstream. The rainfall event occurs in the upstream area, and the impact of floods on the downstream area is predicted. The upstream water level: 2.3 meters, the downstream water level: 1.5 meters, the upstream flow velocity: 1.2 meters / second, the downstream flow velocity: 0.8 meters / second, and the rainfall: 10 mm / hour.

[0077] The long-term trend of the upstream water level was found to be a slow rise of 0.05 m / h, and due to the influence of rainfall, the upstream water level increased by 0.1 m / h in the short term. Spatially mapping the upstream and downstream water levels, it was found that the time lag of water flow from upstream to downstream was 2 hours and the path length was 10 km. The current rainfall is 10 mm / h, but the rainfall intensity is expected to increase to 30 mm / h in the future, which will significantly increase the speed of water level rise, causing the downstream water level to rise rapidly in the next 6 hours.

[0078] The basin is divided into 100 small units, and the analysis shows that the upstream water level will increase from 2.3 meters to 2.8 meters in the next 2 hours, and the downstream water level will increase from 1.5 meters to 2.0 meters in the same period. Calculations show that the upstream water flow rate increases to 1.5 meters per second, and the downstream water flow rate increases to 1.1 meters per second. The flood peak is expected to reach the downstream within 3 hours, when the downstream water level will rise to 2.3 meters. Through hydrodynamic simulation, it is predicted that in the next 6 hours, due to continuous rainfall, the upstream water level will rise to 3.1 meters, and the downstream water level will reach 2.7 meters, exceeding the warning water level.

[0079] The system generates early warning data based on the water situation forecast model. As the downstream warning water level is 2.5 meters, the system issues a flood warning and notifies relevant departments to take emergency measures.

[0080] See also Figures 1 to 4 The present application provides a method for real-time water situation forecasting, comprising the following steps:

[0081] Step S1: obtaining real-time hydrological data;

[0082] In one embodiment, real-time hydrological data can be obtained through sensors (such as water level gauges, rain gauges, etc.) deployed in water bodies such as rivers, lakes, and reservoirs. These sensors transmit data to a central control system via a network or satellite transmission system. Real-time hydrological data includes rainfall, river water level, water flow velocity, flow, etc. Data collection is performed every minute, every hour, or at longer time intervals as needed.

[0083] Step S2: performing real-time hydrological data analysis based on the real-time hydrological data to obtain real-time hydrological analysis data;

[0084] In one embodiment, by comparing historical hydrological data with current real-time data, the changing trends of current water level, flow and other indicators are analyzed. Comprehensive analysis is performed on various indicators to determine whether there are abnormal conditions, such as rapid rise in water level, abnormal rainfall intensity, etc. The status of the current hydrological data is obtained based on the analysis, and real-time hydrological analysis data is generated.

[0085] Step S3: constructing a water regime forecast model based on real-time hydrological analysis data to obtain a water regime forecast model;

[0086] In one embodiment, parameters are selected based on real-time hydrological analysis data to build a model. Parameters include factors such as rainfall intensity, rainfall duration, historical water level data, and basin area. Based on these parameters, the development trend of water conditions in the future period of time is estimated, such as changes in water levels, increases or decreases in flow, etc. The core of the model is to simulate how rainfall is transmitted to the downstream through the water system, resulting in changes in the water levels of rivers and reservoirs. The model includes considerations of factors such as the gradual accumulation process of rainfall, soil moisture content, and evaporation. After the model is initially established, it can be optimized through repeated iterations and historical data verification to make it more accurate. Real-time data is input into the model in each iteration and compared with the actual results to adjust the model parameters.

[0087] In one embodiment, based on real-time hydrological analysis data, parameters are selected to establish a forecast model. The parameters include rainfall, basin area, hydrogeological conditions, evaporation, etc. The key to the model is to predict the changes in water level and flow in the future based on rainfall and current water level and flow. These parameters are used for calculation. For example, based on the current rainfall of 15 mm and the water level rise rate of 0.3 m / h, the water level changes in the next 3 hours are predicted. At the same time, the basin area and hydrological historical trends are used to estimate the future changes in river flow.

[0088] Step S4: Generate hydrological warning data according to the water regime forecast model to obtain hydrological warning data for real-time water regime forecasting.

[0089] In one embodiment, hydrological warning data is generated in real time based on the constructed water situation forecast model, including predicted water level, flow and other hydrological indicators at future times. If it is predicted that the water level in certain areas may exceed the warning line, the system will automatically issue a warning. The warning data includes information such as the predicted maximum water level, flow, and time of flood arrival. The warning level is divided into yellow warning, orange warning or red warning according to the change range and trend of the water level, prompting relevant managers to take action. The generated warning data will be notified to relevant departments and the public through predetermined channels (such as SMS, APP, broadcast, etc.) so that they can take corresponding preventive measures.

[0090] In one embodiment, an early warning message is generated based on the water level and flow rate predicted by the model. For example, if the predicted water level exceeds 3 m and the flow rate exceeds , the system can automatically send out early warning signals to remind management departments and the public to take protective measures. According to different water level and flow standards, the early warning level is divided. For example, a water level exceeding 3 m triggers a yellow warning, while a water level exceeding 3.5 m triggers a red warning. The early warning information is sent to relevant departments and the public through SMS, broadcast or other systems.

[0091] Preferably, step S1 specifically includes:

[0092] Step S11: collecting hydrological data through a hydrological monitoring device preset at a first geographical location to obtain first real-time hydrological data;

[0093] In one embodiment, hydrological monitoring equipment, such as water level meters, flow meters, rain gauges, etc., are pre-arranged at a first geographical location (e.g., an upstream river section or a lake). These devices have stable monitoring capabilities and can obtain hydrological parameters regularly. The equipment monitors the hydrological conditions of the current area in real time, such as water level, flow, rainfall, etc. Assume that the hydrological data collected at this location include: water level: 2.2 m, flow: , rainfall: 12 mm.

[0094] Step S12: collecting hydrological data through a hydrological monitoring device preset at a second geographical location to obtain second real-time hydrological data;

[0095] In one embodiment, another hydrological monitoring device is arranged at a second geographical location (e.g., downstream area, diversion outlet). This device has similar functions to the first device and is responsible for collecting local hydrological data, such as water level, flow, rainfall, etc. The second device also collects hydrological data in real time. The collected data at the second geographical location are as follows: water level: 1.8 m, flow: , rainfall: 8 mm.

[0096] Step S13: Integrate the first real-time hydrological data and the second real-time hydrological data to obtain real-time hydrological data.

[0097] In one embodiment, the data of the first and second geographic locations are transmitted to a central server or data processing center via wireless communication (such as 4G / 5G, satellite communication) or wired communication (such as optical fiber). An encryption protocol is used to ensure data security during transmission. During the transmission process, it is necessary to ensure that the data of the first and second geographic locations can be synchronously processed within a unified time range. If there is a network delay or inconsistent data collection time, the data needs to be timestamped and time sequence correction is performed during subsequent processing.

[0098] First, the transmitted hydrological data of the first and second geographical locations are time-synchronized. For example, if the data is collected from different time points or there is a slight delay, the system will align the data according to the timestamp. The water level, flow, rainfall and other data at different times can be aligned by linear interpolation. The hydrological data at different locations are weighted and integrated according to their importance, distance, basin characteristics, etc. For example: if the first geographical location is located in the upstream area, its rainfall data has a greater impact on the overall basin, so it can be given a higher weight. If the second geographical location is in the downstream area, its flow data has a greater impact on the overall hydrological forecast. The data of the first and second geographical locations are integrated to generate unified real-time hydrological data. At this time, the real-time hydrological data contains water level, flow and rainfall information from different geographical locations, and the time difference has been processed to form a complete data set.

[0099] Preferably, step S2 specifically includes:

[0100] Step S21: constructing a hydrological correlation matrix according to the first real-time hydrological data and the second real-time hydrological data to obtain hydrological correlation matrix data;

[0101] In one embodiment, first, real-time hydrological data of the first and second geographical locations are needed. The hydrological data of each location may include indicators such as water level, flow, and rainfall. The various hydrological data items (such as water level, flow, and rainfall) of the first and second geographical locations are used as rows and columns of a matrix. Create an association matrix in which each element represents the correlation or linkage between a certain hydrological data between the two locations. Calculate the correlation between each hydrological data item at the first location and the second location. For example, the correlation strength between the water level at the first location and the water level at the second location is quantified by a correlation coefficient (such as a Pearson correlation coefficient). The elements of the association matrix are the correlation coefficients of different hydrological data items. Fill the correlation data between different hydrological indicators into the matrix.

[0102] Step S22: Performing geographical location factor analysis based on the hydrological correlation matrix data to obtain hydrologically correlated geographical location analysis data;

[0103] In one embodiment, the constructed hydrological association matrix is ​​used to analyze the geographical features of two geographical locations. Geographical location factors may include river slope, distance, terrain differences, soil permeability, etc. Generally speaking, the farther the geographical location is, the lower the correlation of hydrological data may be. Considering geographical distance as an influencing factor, the degree of influence of different geographical locations on hydrological data is calculated. For example, a distance attenuation coefficient is used to adjust the correlation data in the hydrological association matrix to reflect the weakening effect of distance on hydrological linkage. If the first position is located in a high slope area and the second position is located in a low-lying area, it may have different effects on the flow transfer speed and the impact of rainfall. It can be corrected in combination with parameters such as geographical slope and terrain. Considering factors such as distance and terrain, the data in the hydrological association matrix is ​​adjusted to obtain a more realistic hydrological correlation geographical location analysis result.

[0104] Step S23: performing location linkage analysis based on the hydrologically associated geographical location analysis data to obtain hydrologically associated geographical linkage analysis data;

[0105] In one embodiment, the linkage between two geographical locations is further analyzed based on the analysis data of the hydrologically associated geographical locations. This linkage reflects the speed and degree of response of another location when the hydrological conditions at one location change. For example, how long will it take for the change in the upstream water level to affect the downstream area, or how much impact the increase in upstream flow will have on downstream floods. Calculate the time of impact of upstream hydrological changes (such as water level, flow) on the downstream. The propagation time of water level rise or flood peak can be calculated by factors such as flow velocity. The linkage strength between locations is quantified by further analyzing the data in the association matrix. For example, if an increase in upstream rainfall causes a rapid increase in downstream flow, it indicates that the linkage is strong. Based on the analysis results, the linkage data between geographical locations is generated. It will be used to evaluate and predict the linkage of water situation changes in different locations.

[0106] Step S24: Generate real-time hydrological analysis data based on the hydrological-related geographic linkage analysis data to obtain real-time hydrological analysis data.

[0107] In one embodiment, a real-time hydrological analysis result is generated based on the location linkage analysis data and the current real-time hydrological conditions. The data will be used to predict future hydrological changes and early warnings. For example, based on the current rainfall and water level changes, the flood risk in the downstream area is predicted. Combined with the real-time hydrological data from upstream and downstream, the water conditions in the current basin are evaluated. Based on the linkage analysis data, the water level and flow changes within a certain period of time in the future are predicted, especially the flood peak, water accumulation and water flow after rainfall. As more real-time data is input, the hydrological analysis results are continuously updated to provide more accurate water situation forecasts. The current upstream water level is 2.5 m, and the flow is 2.5 m. , rainfall 20 mm. Combined with linkage analysis, the system predicts that the downstream water level will rise by 0.3 m to 2.1 m in the next 4 hours, and the flow will increase to Real-time hydrological analysis data will output these prediction results for subsequent early warning and disaster prevention decisions.

[0108] Preferably, the hydrological correlation matrix is ​​constructed as follows:

[0109] Extracting hydrological characteristics according to the first real-time hydrological data and the second real-time hydrological data to obtain first real-time hydrological characteristic data and second real-time hydrological characteristic data;

[0110] In one embodiment, it is necessary to obtain real-time hydrological data, including water level, flow, rainfall, etc., from hydrological monitoring devices at a first geographical location and a second geographical location. Extract the water level and flow change rates at different time points (for example, the magnitude of the water level rise / fall in the past hour). These change rates can be used as one of the hydrological characteristics. Extract the distribution pattern of hydrological data at different geographical locations. For example, whether the water level changes regularly with different geographical locations. Extract basic statistical indicators, such as maximum, minimum, mean, rate of change, etc., to describe the distribution of hydrological characteristics. First location: water level 2.5 m, flow , rainfall 20 mm, rate of change +0.3 m / h. Second position: water level 1.8 m, flow , rainfall 15 mm, change rate +0.2 m / h. The first real-time hydrological characteristic data: water level change rate +0.3 m / h, flow , rainfall 20 mm. Second real-time hydrological characteristic data: water level change rate +0.2 m / h, flow , rainfall is 15 mm. The extracted characteristic data are converted into structured data for analysis to generate first real-time hydrological characteristic data and second real-time hydrological characteristic data.

[0111] Constructing a terrain undulation correlation matrix according to the first real-time hydrological characteristic data and the second real-time hydrological characteristic data to obtain real-time hydrological terrain undulation correlation matrix data;

[0112] In one embodiment, in combination with the terrain information of the first and second geographical locations, terrain data can be obtained through satellite remote sensing, geographic information system (GIS), etc., including slope, elevation difference, riverbed shape, etc. The impact of terrain undulation on hydrological processes is very significant, especially in the changes of water level and flow. By analyzing the impact of terrain undulation on hydrological data, a terrain undulation association matrix is ​​established. For example, areas with larger slopes may increase the speed of water flow, thereby affecting flow changes. The elements in the matrix represent the degree of influence of the terrain undulation of the two geographical locations on the hydrological characteristics. For example, if the slope of the upstream area is larger, the impact on the downstream water level and flow changes may be greater. In combination with the hydrological characteristic data of the first and second locations, a terrain undulation association matrix is ​​generated to reflect the effect of terrain characteristics on hydrological processes.

[0113] Constructing a watershed distribution association matrix according to the first real-time hydrological characteristic data and the second real-time hydrological characteristic data to obtain real-time hydrological watershed distribution association matrix data;

[0114] In one embodiment, the basin distribution data includes the flow direction, basin area, watershed, catchment area, etc. of the water system. The distribution of the basin has a direct impact on the hydrological process, especially the accumulation effect of flow and rainfall. The distribution characteristics of the hydrological data in the basin are analyzed, especially the convergence and dispersion patterns of water flow. The shape, area, and convergence speed of precipitation of the basin will affect the changes in water level and flow. Each matrix element reflects the impact of basin characteristics on hydrological data. For example, rainfall in areas with larger basin areas is more likely to converge into flows, resulting in increased water levels and flows downstream. Combining the hydrological characteristics of the first and second positions, an association matrix reflecting the impact of basin distribution is generated.

[0115] The initial hydrological correlation matrix data is obtained by performing correlation calculation based on the real-time hydrological terrain fluctuation correlation matrix data and the real-time hydrological watershed distribution correlation matrix data;

[0116] In one embodiment, the terrain undulation correlation matrix and the watershed distribution correlation matrix are combined. The two matrices are integrated by weighted average or other comprehensive methods to obtain a more comprehensive initial hydrological correlation matrix. Combined with the different effects of terrain undulation and watershed distribution on hydrological characteristics, the comprehensive correlation between various hydrological characteristics is calculated. According to the different influence weights of terrain and watershed distribution, the correlation between characteristics such as water level, flow, and rainfall is calculated. The weight is adjusted according to the actual situation. For example, in areas with large terrain changes, the terrain has a greater impact. After completing the correlation calculation, a preliminary hydrological correlation matrix is ​​obtained, which contains the correlation data of characteristics such as water level, flow, and rainfall.

[0117] The initial hydrological correlation matrix data is subjected to correlation matrix sparse processing to obtain the hydrological correlation matrix data.

[0118] In one embodiment, the purpose of the sparse processing is to filter out the weakly correlated elements (correlation below a certain threshold) in the initial hydrological correlation matrix to reduce unnecessary data noise and simplify the model. A threshold is set, for example, elements with a correlation below 0.5 will be set to 0, indicating that the correlation between these data is weak and is not taken into consideration. The elements in the initial matrix are compared with the set threshold, and the elements below the threshold are set to 0, and those above the threshold are retained. After the sparse processing, a hydrological correlation matrix is ​​generated, retaining the characteristic data with high correlation.

[0119] Preferably, the geographical location factor analysis is specifically as follows:

[0120] Conduct terrain and topography impact analysis based on hydrological correlation matrix data to obtain terrain impact data;

[0121] In one embodiment, according to the hydrological association matrix constructed previously, hydrological data such as water level, flow, and rainfall are used as input. The terrain data is used to analyze the slope differences in different geographical locations. When the slope is large, the water flow speed will increase, resulting in increased flow and intensified water level fluctuations. Through the elevation data, the elevation differences in different regions are calculated. The elevation difference will directly affect the water level changes in the hydrological process, especially between the upstream and downstream of the basin. The influence of slope and elevation on hydrological characteristics is analyzed using a numerical elevation model (DEM). The flow rate of water and its influence on flow and water level are estimated through the slope and elevation differences. According to the terrain parameters such as slope and elevation, each feature in the hydrological association matrix (such as water level, flow, etc.) is corrected to generate terrain impact data. The first position: slope 15%, elevation 500 meters, the terrain has a greater impact. The second position: slope 10%, elevation 200 meters, the terrain has a relatively small impact. Through the slope and elevation analysis, the degree of influence of the two places on the water level and flow is determined, for example: the water level changes faster in the uphill area, and the flow increases more significantly. The generated terrain impact data include the water level impact coefficient at the first location: 0.7 (large terrain impact), and the flow impact coefficient at the second location: 0.6 (small terrain impact).

[0122] Conduct soil and vegetation type analysis based on terrain impact data to obtain soil and vegetation impact data;

[0123] In one embodiment, the terrain impact data is input, combined with the terrain impact data generated in the previous step, including the influence coefficients of water level and flow. Different types of soil have different water permeability. For example, sand has high permeability, which leads to rapid water infiltration after rainfall, while clay has poor permeability, which easily leads to increased surface runoff. Analyze the water retention capacity of the soil. Soil with poor permeability is prone to produce more surface runoff, affecting the flow and water level changes (based on database query, and calculate the permeability, water retention rate and surface water leakage rate through expert engine or empirical data). Vegetation helps to slow down surface runoff and increase the infiltration of precipitation. In areas with dense vegetation, the water flow rate is slow after rainfall, and the flow change range is small. The transpiration of vegetation has a greater impact on hydrological characteristics, especially in areas with dense vegetation. It can reduce water backflow and reduce the rate of water level rise. The terrain impact data is corrected by factors such as the permeability of soil type and vegetation coverage, and soil and vegetation impact data are generated. First position: the soil is clay with low permeability; the vegetation coverage rate is 70%, which has a greater impact. Second position: the soil is sandy and has high permeability; the vegetation coverage rate is 40%, and the impact is relatively small. The generated soil and vegetation impact data, first position: soil permeability 0.4, vegetation impact coefficient 0.6 (vegetation has a greater impact on flow). Second position: soil permeability 0.7, vegetation impact coefficient 0.3 (vegetation has a smaller impact on flow).

[0124] Meteorological factors are analyzed based on soil and vegetation impact data to obtain hydrologically related geographical location analysis data.

[0125] In one embodiment, data such as soil permeability and vegetation coverage are used as the basis for meteorological factor analysis. The rainfall intensity in the analysis area. Heavy rainfall will rapidly increase water level and flow, especially in areas with poor permeability soil. Rainfall data can be obtained from meteorological stations or satellite remote sensing data. The evaporation rate will affect the retention capacity of water bodies. High temperatures and strong solar radiation will increase the evaporation rate of water bodies, causing the water level to drop. Wind speed has a significant impact on rainfall and water evaporation, and the evaporation rate is accelerated in areas with higher wind speeds. When humidity is low, precipitation cannot quickly form effective runoff, but is absorbed by vegetation or soil. Combine meteorological data with previous soil and vegetation impact data to adjust the various features in the hydrological association matrix. By analyzing the impact of meteorological factors such as rainfall and evaporation on hydrological data, the correlation between water level, flow, and rainfall is further corrected. Combined with the results of meteorological factor analysis, hydrologically associated geographic location analysis data is generated.

[0126] Preferably, the location linkage association is specifically:

[0127] Perform time-lag correlation analysis based on hydrologically associated geographical location analysis data to obtain hydrologically associated time-lag data;

[0128] In one embodiment, a previously generated hydrological association matrix is ​​first obtained, which includes the correlation of hydrological characteristics of different geographical locations such as water level, flow, and rainfall. The possibility of water flow acceleration or slowdown is determined by slope analysis. Areas with larger slopes will accelerate water flow, causing the downstream water level to rise rapidly, increasing the risk of floods. The altitude differences of different geographical locations are analyzed through elevation data. The water level in high-altitude areas changes faster, and low-altitude areas accumulate more water, forming floods or runoff. According to the terrain characteristics, the correlation between hydrological characteristics is adjusted. For example, in areas with larger slopes, the correlation between flow changes and water level changes may be stronger, so the correlation of these features can be improved in the hydrological association matrix. The analyzed data will generate a matrix for representing the impact of terrain in different geographical locations, reflecting the correction results of terrain factors such as slope and elevation on hydrological characteristics.

[0129] The location transmission model is constructed based on the hydrological location time-lag correlation data to obtain the hydrological transmission model;

[0130] In one embodiment, the terrain impact data generated in the first step is used as input, including the correction results of water level, flow, rainfall and other characteristics. Soil permeability determines the speed of water infiltration. Soil with high permeability will reduce the amount of surface runoff, while soil with poor permeability (such as clay) will increase the risk of runoff and flooding. Soil with strong water retention capacity (such as loam) will slow down the rate of water level rise and reduce the risk of short-term floods. High vegetation coverage will slow down surface runoff and increase water infiltration, thereby reducing the sharp fluctuations in flow and water level. The transpiration of vegetation will consume part of the water, reduce the amount of water flowing into the river, and thus slow down the rise of water level. According to factors such as soil permeability and vegetation coverage, the correlation between hydrological characteristics is further corrected. For example, the water level changes more slowly in areas with high vegetation coverage, and the flow change is also buffered. Combined with the characteristic data of soil and vegetation, the terrain impact data is secondarily corrected to obtain more accurate hydrological characteristic correlation data. First position: soil type: clay, low permeability, high water retention capacity, vegetation coverage: 70%. Second position: Soil type: sandy soil, high permeability, low water retention capacity, vegetation cover: 40%.

[0131] The linkage intensity is processed according to the hydrological transmission model to obtain the hydrologically associated geographic linkage analysis data.

[0132] In one embodiment, the soil and vegetation impact data generated in the second step are used as input in combination with the correction results of the hydrological characteristics. The rainfall intensity and frequency in different time periods are analyzed. The water level rises faster in areas with heavy rainfall, and areas with poor permeability are more likely to produce surface runoff, resulting in increased flow. The evaporation rate is analyzed through meteorological conditions such as temperature, humidity and wind speed. In areas with high evaporation rates, the water level rises more slowly. High temperatures and strong winds accelerate evaporation, reduce the amount of surface water, and affect the conversion rate of precipitation. The correlation of hydrological characteristics is corrected through meteorological data. For example, in areas with high rainfall intensity, the correlation between water level and flow will be significantly improved; in areas with high evaporation rates, the correlation between rainfall and water level may be lower. Combined with the correction of meteorological factors, hydrologically associated geographic location analysis data is generated as a basis for water situation forecasting and risk management.

[0133] Preferably, the linkage strength processing is specifically as follows:

[0134] The mixed linkage intensity is processed according to the hydrological transmission model to obtain the first hydrologically associated geographic linkage analysis data;

[0135] In one embodiment, a transmission model is used, which describes the propagation of hydrological characteristics between different locations, including propagation speed, time and intensity. Hybrid linkage strength refers to calculating the linkage between locations of multiple factors (such as water level, flow, rainfall, etc.). By combining these factors, a linkage strength value is obtained. The hybrid processing method can combine the linkage strength of different hydrological characteristics into a comprehensive value through weighted average, least squares method and other techniques. Generate the first hydrological-related geographic linkage analysis data, and combine the linkage of multiple hydrological characteristics such as water level and flow to generate preliminary linkage strength analysis data. The linkage strength from the first position to the second position: water level linkage strength: 0.9, flow linkage strength: 0.8, rainfall linkage strength: 0.7, through weighted processing, the hybrid linkage strength is: hybrid linkage strength = 0.5×0.9+0.3×0.8+0.2×0.7=0.83. The generated first hydrological-related geographic linkage analysis data is 0.83, reflecting the linkage strength under multiple factors.

[0136] Generate multi-level hysteresis coefficients according to the hydrological transmission model to obtain hysteresis coefficient data;

[0137] In one embodiment, based on the transmission model, the time lag information of hydrological characteristics between different locations is determined. Multi-level lag coefficients refer to analyzing the lag effects of hydrological characteristics in multiple time periods. For example, when rainfall occurs upstream, it will gradually affect the flow and water level downstream within a few hours. By hierarchically analyzing the lag coefficients of different time periods, a more accurate hydrological propagation process can be obtained. Different time levels (for example, 1 hour, 3 hours, 6 hours, 12 hours) can reflect the response of hydrological characteristics in different time periods. Based on the lag effects of different time periods, lag coefficient data are generated to reflect the temporal change pattern of hydrological characteristics.

[0138] In one embodiment, more importantly, multi-level time series segmentation is performed according to the hydrological transmission model to obtain multi-level time series segmentation data of hydrological transmission; periodicity identification is performed according to the multi-level time series segmentation data of hydrological transmission to obtain multi-level periodic data of hydrological transmission; spatial weight adjustment of terrain and watershed distribution is performed according to the multi-level periodic data of hydrological transmission to obtain periodic spatial adjustment data of hydrological transmission; nonlinear lag period detection is performed according to the periodic spatial adjustment data of hydrological transmission to obtain nonlinear lag period detection data; Kalman filter optimization is performed according to the nonlinear lag period detection data to obtain lag period optimization data; coordinated adjustment is performed according to the lag period optimization data to obtain lag period data.

[0139] Perform time series analysis on historical and real-time hydrological data, and divide the hydrological data into different levels according to the time series. Each level represents a specific change pattern in a period, such as a stable period, a rising period, and a fluctuating period. According to the hydrological transmission model, decompose the long, medium, and short-term trends in the hydrological data, and determine the boundary points of these time series segments. Obtain hydrological time series data (such as water level and flow rate). According to the set time window, perform preliminary trend detection on the data to identify the time period changes at each level. Analyze the characteristic values ​​of each time period, such as trend slope, water level change rate, etc., to further refine the segmentation. Generate multi-level time series segmented data for hydrological transmission and output the time range of each level.

[0140] Perform periodic pattern recognition on segmented time series data. By analyzing the hydrological data in each time period, identify the periodic behavior of the data, such as periodic changes in water levels and flood cycles. Use frequency domain analysis technology to convert time domain data into frequency domain data to discover potential periodic patterns. Obtain segmented hydrological data and perform periodic fluctuation analysis. Convert time domain data into frequency domain data to detect periodic patterns. Calculate period amplitude, period duration, and repetition frequency to identify periodic characteristics. Generate multi-level periodic data for hydrological transmission and output the periodic characteristics of each segment.

[0141] Based on the identified periodic data, terrain features (such as slope, elevation) and watershed distribution features (such as watershed area, surface cover) are used as weight factors for adjustment. A weight matrix is ​​constructed to reflect the weight value of each location in spatial transmission. By combining geographic information system (GIS) data, the impact of water flow at different locations is calculated and the spatial association of periodic data is adjusted. The terrain data (such as slope, terrain undulation) and watershed distribution data of each geographical location are obtained. According to the hydrological transmission path, a spatial weight matrix is ​​constructed to determine the priority of water flow transmission at each location. The periodic data is combined with the weight matrix to adjust the impact of the periodic data at different locations. Generate the spatial adjustment data of hydrological transmission periodicity and output the adjusted spatial impact weight.

[0142] Use nonlinear regression analysis to identify hysteresis between different locations. Find the transmission time lag of hydrological changes between upstream and downstream or other geographical locations by fitting nonlinear relationships in hydrological data, especially by detecting nonlinear response relationships in extreme events such as floods. Obtain spatially adjusted hydrological periodicity data. Use nonlinear fitting methods to detect time lag characteristics of upstream and downstream hydrological changes. Identify nonlinear relationships in hysteresis, especially hysteresis effects in sudden events of change. Generate nonlinear hysteresis test data and output hysteresis and nonlinear relationships for each location.

[0143] The nonlinear hysteresis coefficient is optimized by using Kalman filter. Kalman filter is used to dynamically correct the hysteresis coefficient. The hysteresis coefficient is corrected by continuously updated real-time data, and the interference of noise and sudden events is removed to obtain a more stable and accurate hysteresis coefficient estimation value. The nonlinear hysteresis coefficient detection data is obtained. The hysteresis coefficient at the previous moment is used to estimate the current hysteresis coefficient. The hysteresis coefficient is dynamically corrected according to the error between the real-time hydrological data at the current moment and the predicted data. The updated hysteresis coefficient is output to generate the hysteresis coefficient optimization data.

[0144] Based on the hysteresis data optimized by Kalman filter, the hysteresis of different locations and different hydrological variables are coordinated and adjusted. The coordinated adjustment ensures that the hysteresis of different locations maintains a reasonable relationship, while taking into account the linkage of different types of hydrological variables (such as water level and flow rate) in the basin to avoid local failure. Obtain the hysteresis data after Kalman filter optimization. Analyze the linkage of hysteresis of different locations and evaluate whether the hysteresis of adjacent locations maintains a reasonable linkage relationship. Adjust unreasonable hysteresis to ensure that the hysteresis of upstream and downstream and between basins is coordinated. Output the adjusted hysteresis data to generate hysteresis estimation results.

[0145] The hydrological transmission model is subjected to time series regression calculation according to the lag period data to obtain the hydrological time series regression model;

[0146] In one embodiment, lagged period data is used as the input of the time series regression model. Time series regression is used to predict the changing trend of hydrological characteristics. By analyzing past data, changes in water level, flow, etc. in a specific time period in the future are predicted. Time series regression can combine past hydrological characteristic values ​​and lag effects based on lagged period data to predict the values ​​of future hydrological characteristics. Regression methods include autoregression (AR), autoregressive moving average (ARMA), autoregressive integrated moving average model (ARIMA), etc., which are used to capture the time dependence of hydrological characteristics. Through time series regression calculation, a regression model that describes the changes of hydrological characteristics over time is generated to predict the changing trend of future hydrological characteristics. Based on the lagged period data, the AR model is used for time series regression: the historical data of upstream water level changes are used to predict the changes in downstream water levels in the next 6 hours. It is predicted that the downstream water level will increase by 0.5 m in the next 3 hours. The generated time series regression model can be described as follows: , is the water level at a future time (or other hydrological parameters / characteristic data), and is the current and previous water level (or other hydrological parameters / characteristics), for Previous water level (or other hydrological parameter / characteristic data).

[0147] The Bayesian network causal relationship processing is carried out according to the hydrological time series regression model to obtain the second hydrological associated geographic linkage analysis data;

[0148] In one embodiment, the generated time series regression model is used as the basis for Bayesian network causal relationship processing. Bayesian network is a probabilistic model used to analyze the causal relationship between variables. Through the Bayesian network, the causal dependency between upstream and downstream hydrological characteristics can be analyzed, such as the probability of the upstream water level affecting the downstream water level. The Bayesian network can be inferred through the conditional probability table (CPT) to determine the conditional dependency of the upstream water level change on the downstream water level change. Generate the second hydrological association geographic linkage analysis data: According to the causal inference of the Bayesian network, generate the second hydrological association geographic linkage analysis data to reflect the causal correlation between hydrological characteristics. The probability that the upstream water level change causes the downstream water level to rise is 80%. The conditional probability of the increase in upstream flow on the downstream water level is 70%. Upstream water level change → downstream water level change [P (downstream water level | upstream water level) = 0.8], upstream flow change → downstream water level change [P (downstream water level | upstream flow) = 0.7].

[0149] Establish a Bayesian network structure and use the upstream and downstream hydrological characteristics as nodes, such as upstream water level and downstream water level. According to the time series regression model, determine the direction of causality, for example, upstream water level changes lead to downstream water level changes. Using the coefficients in the regression model, calculate the conditional probability of upstream water level changes on downstream water level changes. Infer the probability distribution of downstream water level changes, given the upstream water level changes, calculate the probability of downstream water level changes. Use regression coefficients as parameters for conditional probabilities. For example, After a lag of 2 hours, the is determined, and the regression coefficient is 1.0. Then the conditional probability P( ∣ ) can be expressed as a deterministic relationship. Generate the second hydrological association geographic linkage analysis data, such as Bayesian network analysis results, including conditional probability and causal relationship.

[0150] The first hydrologically associated geographic linkage analysis data and the second hydrologically associated geographic linkage analysis data are adjusted based on similarities and differences to obtain hydrologically associated geographic linkage analysis data.

[0151] In one embodiment, the first hydrological linkage analysis data generated by the hybrid linkage strength processing is compared and integrated with the second hydrological linkage analysis data generated by the Bayesian network causal relationship processing. By comparing the first and second linkage analysis data, the differences and consistencies between different data are determined. For example, if the linkage obtained by the hybrid linkage strength processing is relatively high, while the Bayesian network inference shows that some relationships are weak, further adjustment is required. Adjustments are made based on the differences to correct the inconsistencies in the model and ensure that the final linkage analysis results are more accurate. Generate hydrologically associated geographic linkage analysis data, integrate the first and second linkage analysis data through a feedback mechanism of similarities and differences, and generate hydrologically associated geographic linkage analysis results. First linkage strength analysis data: the hybrid linkage strength is 0.83. Second linkage strength analysis data: the Bayesian network linkage strength is 0.8.

[0152] Preferably, the real-time hydrological analysis data generation is specifically as follows:

[0153] Decompose the time series characteristics according to the hydrologically associated geographic linkage analysis data to obtain the hydrologically associated time series characteristics decomposition data;

[0154] In one embodiment, the previously generated hydrological-related geographic linkage analysis data is used, including the linkage relationship between hydrological characteristics such as water level, flow, rainfall, etc. at different locations. The characteristics of different time periods in the hydrological data are extracted, such as trend terms, periodic terms, random terms, etc. Decomposition methods include empirical mode decomposition (EMD), Hilbert-Huang transform (HHT), etc. The long-term trend and short-term fluctuations of the hydrological characteristics are separated to identify the changes in the long-term trend and the impact of short-term fluctuations. The decomposed data contains the trends and fluctuations of each hydrological characteristic over time.

[0155] According to the hydrological correlation time series feature decomposition data, dynamic spatial mapping is performed to obtain hydrological correlation spatial reconstruction data;

[0156] In one embodiment, the decomposed trend term, period term and random term data are used as input. The time series characteristics are mapped to the hydrological location in the geographic space to determine the spatial changes in the hydrological characteristics of each location. The dynamic spatial mapping technology can estimate the hydrological characteristics of unobserved locations through interpolation, Kriging and other methods to generate a hydrological data distribution map in space. Through dynamic mapping, spatial reconstruction data containing the distribution and change trends of hydrological characteristics in different geographical locations is generated.

[0157] Embedding local meteorological factors according to the hydrological correlation spatial reconstruction data to obtain local meteorological factor embedded data;

[0158] In one embodiment, spatially reconstructed data is used as input, including distribution data of hydrological features at different locations. Local meteorological factors that affect hydrological features, such as rainfall, evaporation, wind speed, etc., are extracted. These factors have a direct impact on the changes in hydrological features. Local meteorological factors are embedded in hydrological features to adjust the values ​​of hydrological features. For example, in high rainfall areas, water levels rise faster. The hydrological feature data after the influence of meteorological factors is integrated to reflect more accurate hydrological changes. Meteorological factor embedding: rainfall: +15 mm / h (upstream), wind speed: 10 km / h, evaporation: -5 mm / h.

[0159] The water flow path model is constructed according to the local meteorological factor embedded data to obtain the water flow path model;

[0160] In one embodiment, hydrological characteristic data embedded with meteorological factors is used as input. The propagation path and speed of water flow are determined by geographical features such as topography and river morphology. Path models include finite difference method, finite element method, etc. The water flow path model is used to simulate the propagation process of water flow from upstream to downstream, and the propagation time lag of water level and flow is calculated. The model can describe the transmission path of water flow between different locations and the changes in its hydrological characteristics.

[0161] Multi-point collaborative processing is performed based on the water flow path model to obtain real-time hydrological analysis data.

[0162] In one embodiment, the constructed water flow path model is used to simulate the propagation of hydrological characteristics among multiple locations. The hydrological characteristic data of multiple locations such as upstream, midstream, and downstream are processed collaboratively to analyze the hydrological changes in the entire basin. The overall changes in hydrological characteristics are calculated in real time based on the hydrological characteristics of each location and the transmission time of the water flow path, generating dynamic hydrological analysis data. The results of multi-point collaborative processing provide real-time hydrological status of each location in the basin.

[0163] Preferably, step S3 specifically includes:

[0164] Step S31: performing finite element analysis based on the real-time hydrological analysis data to obtain real-time hydrological finite element data;

[0165] In one embodiment, the real-time hydrological analysis data generated in the previous step is used, including the water level, flow, rainfall and other characteristics of each point in the basin. The basin is divided into multiple units using the finite element method, and the changes in hydrological characteristics are analyzed point by point. The finite element method can accurately simulate the hydrological changes in the entire area by solving the hydrological equations on each unit. The basin is divided into several finite element units (grids). Each unit represents a small area, and the hydrological changes in the entire basin are estimated by calculating each unit. The water level and flow in each grid are solved to obtain the changes in the hydrological characteristics of each small area. The result of the finite element analysis is the distribution of hydrological characteristics in the entire basin, including the changes in water level and flow in each grid, which reflects the spatial changes in the hydrological characteristics of the basin.

[0166] Step S32: performing hydrodynamic calculations based on the real-time hydrological finite element data to obtain real-time hydrological and hydrodynamic data;

[0167] In one embodiment, the hydrological characteristics of each grid obtained by finite element analysis include information such as water level, flow, and pressure. The dynamic characteristics of the water flow, such as velocity, pressure, and momentum, are calculated through a hydrodynamic model. Hydrodynamic calculations are based on fluid mechanics equations, such as the Navier-Stokes equations, which are used to simulate the movement of water in a watershed. Hydrodynamic calculations are performed on each grid to solve key parameters such as the velocity field and pressure field of the water flow. Considering the viscosity, inertia, and other characteristics of the water flow, the movement laws of the water flow under different terrains are analyzed. Through hydrodynamic calculations, the hydrodynamic characteristics of each grid are obtained, reflecting the dynamic changes of the water flow velocity, pressure, and momentum.

[0168] Step S33: constructing a water regime forecast model based on real-time hydrological and hydrodynamic data to obtain a water regime forecast model.

[0169] In one embodiment, dynamic hydrological characteristics such as flow rate and pressure obtained by hydrodynamic calculation are used as input. Based on the hydrodynamic data, a water situation forecast model is constructed to predict future water levels, flows and flood risks. The water situation forecast model can be based on a physical model, a statistical model or a machine learning model, combined with current hydrological and meteorological conditions, to predict future hydrological changes. Physical model: prediction based on hydrological and hydrodynamic equations. Statistical model: regression prediction using historical hydrological data. Machine learning model: prediction using algorithms such as deep learning or decision trees combined with big data. Training and calibration with historical hydrological data optimizes the prediction accuracy of the model. The model can predict future hydrological changes, including water levels, flows, flood warnings and other content, based on current hydrological conditions.

[0170] Preferably, the present application also provides a water condition real-time forecasting system for executing the water condition real-time forecasting method as described above, the water condition real-time forecasting system comprising:

[0171] Real-time hydrological data acquisition module, used to obtain real-time hydrological data;

[0172] A real-time hydrological data analysis module is used to perform real-time hydrological data analysis based on real-time hydrological data to obtain real-time hydrological analysis data;

[0173] A water regime forecast model construction module is used to construct a water regime forecast model based on real-time hydrological analysis data to obtain a water regime forecast model;

[0174] The hydrological warning data generation module is used to generate hydrological warning data according to the water situation forecast model to obtain hydrological warning data for real-time water situation forecasting operations.

[0175] Therefore, from any point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the attached application documents rather than the above description, and it is intended that all changes falling within the meaning and scope of equivalent elements of the application documents are included in the present invention.

[0176] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for real-time water situation forecasting, characterized in that: The following steps are involved: Step S1: obtaining real-time hydrological data; Step S2: performing real-time hydrological data analysis based on the real-time hydrological data to obtain real-time hydrological analysis data; Step S3: constructing a water regime forecast model based on real-time hydrological analysis data to obtain a water regime forecast model; Step S4: Generate hydrological warning data according to the water regime forecast model to obtain hydrological warning data for real-time water regime forecasting; The specific step S1 is: Collecting hydrological data through a hydrological monitoring device preset at a first geographical location to obtain first real-time hydrological data; Collecting hydrological data through a hydrological monitoring device preset at a second geographical location to obtain second real-time hydrological data; Integrate the first real-time hydrological data and the second real-time hydrological data to obtain real-time hydrological data; Wherein step S2 is specifically: Constructing a hydrological correlation matrix according to the first real-time hydrological data and the second real-time hydrological data to obtain hydrological correlation matrix data; Perform geographic location factor analysis based on hydrological correlation matrix data to obtain hydrological correlation geographic location analysis data; Perform location linkage analysis based on hydrologically associated geographical location analysis data to obtain hydrologically associated geographical linkage analysis data; Generate real-time hydrological analysis data based on hydrologically associated geographic linkage analysis data to obtain real-time hydrological analysis data; The specific location linkage relationships are: Perform time-lag correlation analysis based on hydrologically associated geographical location analysis data to obtain hydrologically associated time-lag data; The location transmission model is constructed based on the hydrological location time-lag correlation data to obtain the hydrological transmission model; The linkage intensity is processed according to the hydrological transmission model to obtain the hydrological-related geographic linkage analysis data; The linkage strength processing is as follows: The mixed linkage intensity is processed according to the hydrological transmission model to obtain the first hydrologically associated geographic linkage analysis data; Generate multi-level hysteresis coefficients according to the hydrological transmission model to obtain hysteresis coefficient data; The hydrological transmission model is subjected to time series regression calculation according to the lag period data to obtain the hydrological time series regression model; The Bayesian network causal relationship processing is carried out according to the hydrological time series regression model to obtain the second hydrological associated geographic linkage analysis data; Performing feedback adjustment on the first hydrologically associated geographic linkage analysis data and the second hydrologically associated geographic linkage analysis data based on similarities and differences to obtain hydrologically associated geographic linkage analysis data; The generation of multi-level lag period data is as follows: Perform multi-level time series segmentation according to the hydrological transmission model to obtain multi-level time series segmentation data of hydrological transmission; According to the multi-level time series segmented data of hydrological transmission, periodicity identification is performed to obtain multi-level periodicity data of hydrological transmission; According to the multi-level periodic data of hydrological transmission, the spatial weight of terrain and watershed distribution is adjusted to obtain the periodic spatial adjustment data of hydrological transmission; Nonlinear hysteresis period detection is performed according to the hydrological transmission periodic spatial adjustment data to obtain nonlinear hysteresis period detection data; Kalman filter optimization is performed according to nonlinear lag period detection data to obtain lag period optimization data; According to the lag period optimization data, coordinated adjustments are made to obtain the lag period data.

2. The method according to claim 1, characterized in that The hydrological correlation matrix is ​​constructed as follows: Extracting hydrological characteristics according to the first real-time hydrological data and the second real-time hydrological data to obtain first real-time hydrological characteristic data and second real-time hydrological characteristic data; Constructing a terrain undulation correlation matrix according to the first real-time hydrological characteristic data and the second real-time hydrological characteristic data to obtain real-time hydrological terrain undulation correlation matrix data; Constructing a watershed distribution association matrix according to the first real-time hydrological characteristic data and the second real-time hydrological characteristic data to obtain real-time hydrological watershed distribution association matrix data; The initial hydrological correlation matrix data is obtained by performing correlation calculation based on the real-time hydrological terrain fluctuation correlation matrix data and the real-time hydrological watershed distribution correlation matrix data; The initial hydrological correlation matrix data is subjected to correlation matrix sparse processing to obtain the hydrological correlation matrix data.

3. The method according to claim 1, characterized in that The analysis of geographical location factors is as follows: Conduct terrain and topography impact analysis based on hydrological correlation matrix data to obtain terrain impact data; Conduct soil and vegetation type analysis based on terrain impact data to obtain soil and vegetation impact data; Meteorological factors are analyzed based on soil and vegetation impact data to obtain hydrologically related geographical location analysis data.

4. The method according to claim 1, characterized in that: The real-time hydrological analysis data generation is specifically as follows: Decompose the time series characteristics according to the hydrologically associated geographic linkage analysis data to obtain the hydrologically associated time series characteristics decomposition data; According to the hydrological correlation time series feature decomposition data, dynamic spatial mapping is performed to obtain hydrological correlation spatial reconstruction data; Embedding local meteorological factors according to the hydrological correlation spatial reconstruction data to obtain local meteorological factor embedded data; The water flow path model is constructed according to the local meteorological factor embedded data to obtain the water flow path model; Multi-point collaborative processing is performed based on the water flow path model to obtain real-time hydrological analysis data.

5. The method according to claim 1, characterized in that: Step S3 is specifically as follows: Conduct finite element analysis based on real-time hydrological analysis data to obtain real-time hydrological finite element data; Perform hydrodynamic calculations based on real-time hydrological finite element data to obtain real-time hydrological and hydrodynamic data; The water regime forecast model is constructed according to the real-time hydrological and hydrodynamic data to obtain the water regime forecast model.

6. A real-time water situation forecasting system, characterized in that: Used to execute the water regime real-time forecasting method as claimed in claim 1, the water regime real-time forecasting system comprises: Real-time hydrological data acquisition module, used to obtain real-time hydrological data; A real-time hydrological data analysis module is used to perform real-time hydrological data analysis based on real-time hydrological data to obtain real-time hydrological analysis data; A water regime forecast model construction module is used to construct a water regime forecast model based on real-time hydrological analysis data to obtain a water regime forecast model; The hydrological warning data generation module is used to generate hydrological warning data according to the water situation forecast model to obtain hydrological warning data for real-time water situation forecasting operations.

Images