AI identification method and device for river flood control and disaster reduction risk points based on digital twin computable engine

Abstract

The present invention discloses an AI identification method and device for river flood control and disaster reduction risk points based on a digital twin computable engine. The method includes: obtaining three-dimensional scene basic data and water conservancy project parameters in the basin; obtaining real-time rainfall conditions, water conditions, soil moisture content, reservoir conditions, and levee monitoring data to obtain real-time data; establishing physical entities and twin simulation scenes based on the three-dimensional scene basic data and water conservancy project parameters; performing AI simulation calculations of water conditions based on the real-time data to obtain simulation results; performing twin computable simulations based on the physical entities, twin simulation scenes, and the simulation results to obtain simulation results; identifying risk points in the river in the basin based on the simulation results. By implementing the method of the present invention, the problem that digital twin technology mainly lacks intelligent risk identification and multi-factor integration in river flood control and disaster reduction, which affects prediction accuracy, can be solved.

Description

Technical Field

[0001] The present invention relates to a method for identifying risk points for flood control and disaster reduction in a river channel, and more specifically to an AI identification method and device for risk points for flood control and disaster reduction in a river channel based on a digital twin computable engine. Background Art

[0002] Climate change and frequent extreme weather events have brought tremendous pressure to flood control in cities and mountainous areas. Therefore, flood control and disaster reduction in river basins have become an important task of water conservancy management. my country has complex terrain and numerous rivers, and faces the threat of natural disasters such as floods and droughts.

[0003] Traditional flood prevention and disaster reduction methods mainly rely on historical data, river monitoring, and statistical analysis of rainfall and flow stations. They are unable to identify potential risk points in the basin in a timely manner, resulting in a lack of effective response measures when floods strike. Flood prevention actions are often taken only after extreme events such as floods overtopping levees or dam breaches, which can easily cause serious casualties and property losses.

[0004] In recent years, the development of digital twin technology has provided new solutions for river flood control and disaster reduction. Digital twin combines physical entities with their digital models, dynamically monitors and predicts actual conditions through real-time data collection and analysis. In the field of water conservancy, digital twins can help managers grasp the water conditions, rainfall conditions, geology, vegetation and other information of rivers in real time, thereby optimizing flood control and disaster reduction decisions.

[0005] However, the current application of digital twin technology in river flood prevention and disaster reduction is still insufficient. First, although there are many cases of digital twin basin construction, most of them focus on data visualization and monitoring, lacking intelligent identification and dynamic assessment of risk points. Second, the failure to effectively integrate multiple factors such as water conditions, rainfall conditions, geology, landforms and water conservancy projects has led to insufficient accuracy in flood risk prediction.

[0006] Therefore, it is necessary to design a new method to solve the problem that digital twin technology mainly lacks intelligent risk identification and multi-factor integration in river flood control and disaster reduction, which affects the prediction accuracy. Summary of the invention

[0007] The purpose of the present invention is to overcome the defects of the prior art and provide an AI identification method and device for river flood control and disaster reduction risk points based on a digital twin computable engine.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: an AI identification method for river flood control and disaster reduction risk points based on a digital twin computable engine, comprising:

[0009] Obtain basic three-dimensional scene data and water conservancy project parameters within the basin;

[0010] Obtain real-time rainfall, water conditions, soil moisture, reservoir conditions, and levee monitoring data to obtain real-time data;

[0011] Establishing physical entities and twin simulation scenes according to the three-dimensional scene basic data and water conservancy project parameters;

[0012] Perform AI simulation calculation of water conditions according to the real-time data to obtain simulation results;

[0013] Performing twin computable simulation according to the physical entity, the twin simulation scenario and the simulation result to obtain a simulation result;

[0014] The risk points of the river in the basin are identified according to the simulation results.

[0015] Its further technical solution is: the basic data of the three-dimensional scene includes river networks, lakes, wetlands, flood discharge areas, topography, geology, soil, vegetation types, buildings and levee information; the water conservancy project parameters include flood discharge area capacity, levee parameters, regional soil characteristics and historical rainfall and flood data.

[0016] A further technical solution is: establishing a physical entity and a twin simulation scene according to the three-dimensional scene basic data and water conservancy project parameters, including:

[0017] Constructing a physical entity model according to the three-dimensional scene basic data;

[0018] Associating the hydraulic engineering parameters with the physical entity model;

[0019] Build twin simulation scenarios;

[0020] The associated physical entity model is integrated into the twin simulation scene to obtain the physical entity and the twin simulation scene.

[0021] A further technical solution is: performing AI simulation calculation of water conditions according to the real-time data to obtain simulation results, including:

[0022] The CREST model is used to simulate the hydrological process of the basin. Through its distributed characteristics, it integrates the flow generation and runoff processes to generate hydrological results.

[0023] The SWAT model is used to quickly analyze soil moisture, runoff and erosion processes to obtain time series analysis results;

[0024] Use the CREST model to simulate the hydrology of the watershed and the SWAT model to analyze the result data of soil moisture, runoff and erosion processes, use machine learning algorithms to predict watershed runoff and surface runoff, and optimize model parameters to improve the accuracy of watershed runoff and surface erosion;

[0025] The hydrological process results, the time series runoff analysis results, and the simulation and prediction of the model modification are combined to finally form the precipitation runoff and runoff simulation results that are consistent with the basin.

[0026] A further technical solution is: performing twin computable simulation according to the physical entity, the twin simulation scenario and the simulation result to obtain a simulation result, including:

[0027] According to the simulation results, using a twin computable simulation engine to simulate surface precipitation and water fluid in the basin, and forming local runoff water fluid according to the physical entity and the twin simulation scenario;

[0028] The physical mechanism algorithm and the local runoff water fluid are combined to simulate the scouring effect of floods on rivers and embankments, forming a real erosion simulation effect on the river channel. The twin scene dynamically updates the simulated erosion, damage, and overflowing effects of rivers and embankments to obtain simulation results.

[0029] A further technical solution is: identifying risk points of a river in a watershed according to the simulation results includes:

[0030] The AI ​​model is used to analyze the river engineering operating parameters in the simulation results to determine the danger level of the eroded river section and identify the risk points of the river in the basin.

[0031] A further technical solution thereof is: after identifying the risk points of the river in the basin according to the simulation results, it also includes:

[0032] The simulation results are combined with the river risk points, and a combined image is output.

[0033] The present invention also provides an AI identification device for river flood control and disaster reduction risk points based on a digital twin computable engine, comprising:

[0034] Basic data acquisition unit, used to obtain basic three-dimensional scene data and water conservancy project parameters in the basin;

[0035] A real-time data acquisition unit is used to obtain real-time rainfall, water conditions, soil moisture content, reservoir conditions, and levee monitoring data to obtain real-time data;

[0036] An establishing unit, used to establish a physical entity and a twin simulation scene according to the three-dimensional scene basic data and water conservancy project parameters;

[0037] A simulation unit, used for performing AI simulation calculation of water conditions according to the real-time data to obtain simulation results;

[0038] A simulation unit, configured to perform a twin computable simulation according to the physical entity, the twin simulation scenario and the simulation result to obtain a simulation result;

[0039] An identification unit is used to identify risk points of the river in the basin according to the simulation results.

[0040] The present invention further provides a computer device, which includes a memory and a processor. The memory stores a computer program, and the processor implements the above method when executing the computer program.

[0041] The present invention also provides a storage medium, wherein the storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0042] The beneficial effects of the present invention compared with the prior art are as follows: the present invention obtains basic data of three-dimensional scenes and water conservancy project parameters in the basin, combines real-time monitoring data such as rainfall conditions, water conditions, and soil moisture content, establishes physical entities and twin simulation scenes based on these data, calculates real-time water conditions through AI simulation, generates simulation results, and finally uses the simulation results to identify risk points in the river channel in the basin, so as to solve the problem that digital twin technology lacks intelligent risk identification and multi-factor integration in river flood control and disaster reduction, which affects the accuracy of prediction.

[0043] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0045] Figure 1 A schematic diagram of an application scenario of an AI identification method for river flood control and disaster reduction risk points based on a digital twin computable engine provided in an embodiment of the present invention;

[0046] Figure 2 A flow chart of an AI identification method for river flood control and disaster reduction risk points based on a digital twin computable engine provided in an embodiment of the present invention;

[0047] Figure 3 A schematic diagram of a sub-process of an AI identification method for river flood control and disaster reduction risk points based on a digital twin computable engine provided in an embodiment of the present invention;

[0048] Figure 4A schematic diagram of a sub-process of an AI identification method for river flood control and disaster reduction risk points based on a digital twin computable engine provided in an embodiment of the present invention;

[0049] Figure 5 A schematic diagram of a sub-process of an AI identification method for river flood control and disaster reduction risk points based on a digital twin computable engine provided in an embodiment of the present invention;

[0050] Figure 6 A flow chart of a method for identifying risk points of river flood control and disaster reduction using an AI model based on a digital twin computable engine provided by another embodiment of the present invention;

[0051] Figure 7 A schematic block diagram of an AI model identification device for river flood control and disaster reduction risk points based on a digital twin computable engine provided in an embodiment of the present invention;

[0052] Figure 8 A schematic block diagram of a building unit of an AI model identification device for river flood control and disaster reduction risk points based on a digital twin computable engine provided in an embodiment of the present invention;

[0053] Fig. 9 A schematic block diagram of a simulation unit of an AI model identification device for river flood control and disaster reduction risk points based on a digital twin computable engine provided in an embodiment of the present invention;

[0054] Fig.10 A schematic block diagram of a simulation unit of an AI model identification device for river flood control and disaster reduction risk points based on a digital twin computable engine provided in an embodiment of the present invention;

[0055] Fig.11 A schematic block diagram of an AI model identification device for river flood control and disaster reduction risk points based on a digital twin computable engine provided by another embodiment of the present invention;

[0056] Fig.12 A schematic block diagram of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0059] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0060] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0061] See also Figure 1 and Figure 2 , Figure 1 Schematic diagram of the application scenario of the AI ​​identification method for river flood control and disaster reduction risk points based on the digital twin computable engine provided in an embodiment of the present invention. Figure 2 A schematic flow chart of an AI identification method for flood control and disaster reduction risk points in rivers based on a digital twin computable engine provided in an embodiment of the present invention. The AI ​​identification method for flood control and disaster reduction risk points in rivers based on a digital twin computable engine is applied in a server. The server interacts with the terminal and, based on the technology of the digital twin computable engine and a variety of factors, can dynamically identify risk points in the river and improve the flood control and disaster reduction capabilities of the basin through twin computable simulation and AI algorithms. The engine combines the physical entities in the basin with its digital model and flood control and water conservancy system, collects and analyzes data in real time, and quickly simulates the surface runoff and soil infiltration of the basin.

[0062] In the digital twin scenario, the technology generates simulated flood fluids, simulating the real-time convergence and underground infiltration of floods in terrain, landforms and river channels. Managers can obtain real-time information on water conditions, rainfall conditions, geology and vegetation in the river channel, and use the twin engine to generate flood fluids based on the converged surface runoff, and analyze the impact of its kinetic energy and potential energy in the basin on the surrounding terrain, including the scouring and confluence of wetlands, artificial lagoons, reservoirs and river channels.

[0063] Combined with AI computing power, the system can timely and dynamically identify risk points such as river overflow, leakage and dam breach, thereby providing support for flood control and disaster reduction decision-making. In addition, the twin computing engine can not only simulate the rainfall and water conditions of the basin in real time, but also evaluate the rationality and safety of river engineering design by simulating different hydrological scenarios, thereby optimizing the design of water conservancy projects and improving the efficiency and accuracy of flood control and disaster reduction.

[0064] Figure 2 Schematic diagram of the process of AI identification method for river flood control and disaster reduction risk points based on digital twin computable engine provided by an embodiment of the present invention. Figure 2 As shown, the method includes the following steps S110 to S160.

[0065] S110, obtaining basic three-dimensional scene data and water conservancy project parameters in the watershed.

[0066] In this embodiment, the basic data of the three-dimensional scene includes river networks, lakes, wetlands, flood discharge areas, topography, geology, soil, vegetation types, buildings and levee information; the water conservancy project parameters include flood discharge area capacity, levee parameters, regional soil characteristics and historical rainfall and flood data.

[0067] The 3D scene basic data is the core of building a 3D computable twin scene of a watershed, including river networks, lakes, wetlands, flood discharge areas, protected areas, digital elevation models (DEMs), geology, soil, vegetation types, buildings and levees, etc. These data provide the necessary information for understanding the natural characteristics and human activities of the watershed.

[0068] The collection and processing of water conservancy and soil flood control engineering data covers the capacity of the flood discharge area, engineering parameters of the river network embankment (such as embankment width and slope conditions), embankment top elevation, characteristic parameters of the flood control reservoir, as well as regional soil moisture content and permeability, vegetation coverage, etc. These data are helpful in evaluating flood control capabilities and soil protection effects.

[0069] In addition, the collection of historical rainfall and flood data is crucial, as it provides a temporal reference for the model and helps analyze the response of the basin under different meteorological conditions. By integrating these basic data, a three-dimensional computable twin scene of the basin can be effectively built and restored, while providing strong basic data support for the training and calculation of large AI models, thereby achieving more accurate risk identification and decision support.

[0070] S120, obtaining real-time rainfall conditions, water conditions, soil moisture content, reservoir conditions, and levee monitoring data to obtain real-time data.

[0071] In this embodiment, various types of sensors and monitoring equipment are deployed in the watershed to collect a variety of environmental data in real time, including rainfall conditions, water conditions, vegetation water conservation, soil moisture content, reservoir capacity changes, and embankment status monitoring data. These data can be combined with local meteorological elements such as meteorological data and satellite cloud images to provide a comprehensive information basis for hydrological analysis of the watershed. Using these real-time data, AI can perform analyses such as precipitation distribution and surface runoff convergence, thereby improving the ability to predict and manage flood risks and supporting more accurate water resource allocation and ecological protection strategies. Through this comprehensive data collection and analysis, it is possible to monitor basin changes in real time and respond to possible natural disasters in a timely manner.

[0072] S130: Establish a physical entity and a twin simulation scene according to the three-dimensional scene basic data and water conservancy project parameters.

[0073] In this embodiment, physical entities refer to specific objects or systems in the real world, such as rivers, lakes, buildings, etc. The twin simulation scene is a digital replica of these physical entities, which can be simulated by computer models and can reflect the status and behavior of the physical entities in real time.

[0074] In one embodiment, see Figure 3 , the above-mentioned step S130 may include steps S131~S134.

[0075] S131. Construct a physical entity model according to the three-dimensional scene basic data.

[0076] In this embodiment, necessary information is extracted from the three-dimensional scene basic data to construct a corresponding physical entity model. These data may include the geometric shapes and attributes of terrain, rivers, buildings, etc.

[0077] By building an accurate physical entity model, we can ensure the accuracy of the model in the real world and lay the foundation for subsequent analysis and simulation.

[0078] S132. Associating the water conservancy project parameters with the physical entity model.

[0079] In this embodiment, parameters related to the water conservancy project (such as flow, levee height, etc.) are associated with the previously constructed physical entity model to ensure that these parameters can be accurately reflected in the model.

[0080] This association makes the physical model more practical, and can take into account the actual impact of water conservancy projects during the simulation process, thus improving the reliability of the simulation results.

[0081] S133. Construct a twin simulation scenario.

[0082] In this embodiment, a virtual twin simulation scene is created using the constructed physical entity model and the associated water conservancy project parameters. This scene can dynamically reflect the state of the physical entity.

[0083] Twin simulation scenarios allow for real-time analysis and prediction, can simulate changes under different conditions, and provide decision support.

[0084] S134. Integrate the associated physical entity model into the twin simulation scene to obtain the physical entity and the twin simulation scene.

[0085] In this embodiment, the associated physical entity model is integrated into the twin simulation scene to form a complete and interconnected system.

[0086] This integration ensures interaction between physical entities and the simulation environment, improving the accuracy and effectiveness of simulations, allowing decision makers to better understand and respond to real-world situations.

[0087] Overall, implementation of these steps can help improve the efficiency and accuracy of watershed management, thereby supporting more scientific water resources planning and management.

[0088] By collecting basic geographic data of the basin, geological simulation models of the river network, the geometric shape of the levee, material properties, and related boundary conditions, a high-precision basin physical entity and twin simulation scene can be constructed. This twin scene is closely associated and uniquely mapped with the physical entity and its basic data, so that the twin model not only has the characteristics of the physical entity, but also can perform effective calculations.

[0089] Such twin scenarios can participate in physical simulation calculations, and with the help of AI's analysis and evolution capabilities, assist in more complex simulation calculations. This combination enables us to deeply understand the dynamic changes of the basin and provide strong support for decision-making.

[0090] S140, performing AI simulation calculation of the water situation according to the real-time data to obtain a simulation result.

[0091] In this embodiment, the simulation results refer to the hydrological results, the analysis results of the time series and the runoff of the watershed.

[0092] In one embodiment, see Figure 4 , the above-mentioned step S140 may include steps S141~S144.

[0093] S141. Use the CREST model to simulate the hydrological process of the basin, integrate the flow generation and runoff processes through its distributed characteristics, and generate hydrological results;

[0094] Specifically, it includes:

[0095] Basic data utilization: Utilize high-resolution meteorological data (such as rainfall, temperature, humidity, etc.) of the twin scene and basic data such as terrain and soil characteristics that can be calculated by the twin. Meteorological data is usually obtained through remote sensing or weather stations, and can also be short-term meteorological forecasts and forecasts.

[0096] Rainfall runoff process: In the CREST model, rainfall is first intercepted by the vegetation canopy, and then the excess precipitation reaches the surface. The model uses the variable infiltration capacity curve (VIC) to divide precipitation into surface runoff and infiltration. The infiltration part further affects soil moisture and groundwater level.

[0097] Runoff simulation: The CREST model uses a multilinear reservoir method to simulate the runoff path between surface and underground grids. When the surface runoff reaches the next grid, the model continues to apply the VIC theory to divide it into infiltration and overland flow. This method takes into account the interaction between different grids and ensures the accuracy of the hydrological process.

[0098] Output results: After a series of calculations, combined with the distributed characteristics of the CREST model, the hydrological elements of each grid can be generated, such as soil moisture, surface runoff, groundwater runoff, etc.

[0099] S142. Use the SWAT model to quickly analyze soil moisture, runoff and erosion processes to obtain time series analysis results;

[0100] Specifically, twin computable data include meteorological data (precipitation, temperature, wind speed, etc.), soil properties (such as soil type, soil moisture), terrain information (digital elevation model), and land use data. These twin computable data will be used to construct the analysis input data of the SWAT model.

[0101] Watershed division and Hydrological Response Unit (HRU) definition:

[0102] The basin is divided into multiple hydrological response units (HRUs) based on the topography and land use characteristics of the basin. Each HRU represents an area with the same land cover, soil type and management measures. The SWAT model simulates hydrological processes through these HRUs, allowing the model to reflect the spatial heterogeneity within the basin.

[0103] Model setup and parameter calibration

[0104] In the SWAT model, relevant parameters are set, including parameters of precipitation, evapotranspiration, runoff generation and erosion. The model is calibrated with historical observation data to ensure that the simulation results are consistent with the actual observation data.

[0105] Simulation run:

[0106] Run the SWAT model to simulate and calculate the time series of soil moisture, runoff, erosion and other results. The model will dynamically calculate the hydrological process of each HRU based on the input meteorological data and set parameters, and generate corresponding output results.

[0107] Result analysis and output:

[0108] The data output by the analysis model include soil moisture changes, runoff and erosion, etc. At the same time, time series data is generated for further analysis of the occurrence, development and change process of runoff.

[0109] S143. Use the CREST model to simulate the hydrology of the watershed and the SWAT model to analyze the result data of soil moisture, runoff and erosion processes. Use machine learning algorithms to predict watershed runoff and surface runoff. At the same time, tune the model parameters to improve the accuracy of watershed runoff and surface erosion.

[0110] S144. Combining the hydrological process results, the time series runoff analysis results, and the simulation and prediction of the model correction, ultimately forming precipitation runoff and runoff simulation results that match the basin.

[0111] In this embodiment, in watershed management, the CREST distributed hydrological model is used in combination with the collected rainfall, water conditions and meteorological data to more accurately simulate the interaction between surface and underground runoff. This model takes into account the runoff generation process and runoff scheme in the watershed, generates high-resolution hydrological simulation results, and improves the accuracy of prediction.

[0112] At the same time, the SWAT model provides rapid analysis of processes such as soil moisture, runoff and erosion, forming multi-time series data, providing important support for hydrological analysis within the basin.

[0113] The accuracy of runoff prediction has been further improved through artificial intelligence, especially machine learning. Using historical data to train a recurrent neural network (RNN) can dynamically capture the temporal characteristics of precipitation processes and runoff changes, forming continuous rainfall data.

[0114] Finally, the rainfall and water flow results generated by the above models and algorithms are compared with the actual monitoring data, and the model is corrected and trained after the deviation is identified. Through repeated calculations and simulations, the prediction accuracy of precipitation process calculations and surface runoff convergence is effectively improved, thus forming highly reliable basin flood simulation and calculation results.

[0115] S150. Perform twin computable simulation according to the physical entity, the twin simulation scenario and the simulation result to obtain a simulation result.

[0116] In this embodiment, the simulation results refer to the dynamic simulation output of the basin hydrological process obtained by the twin computable simulation engine, including the behavior of water flow, the impact of floods on river channels and levees, etc. These results help to understand and predict flood behavior and its impact on the environment.

[0117] In one embodiment, see Figure 5 , the above-mentioned step S150 may include steps S151~S152.

[0118] S151. Based on the simulation results, the twin computable simulation engine is used to simulate the surface precipitation and water fluid in the basin, and local runoff water fluid is formed based on the physical entity and the twin simulation scenario.

[0119] In this embodiment, a twin computable simulation engine is used to model surface precipitation in the basin based on simulation results collected in the early stage; in this process, local runoff water fluid is formed, taking into account physical factors such as topography and landforms.

[0120] Through physical modeling, the dynamic changes of precipitation and water flow can be simulated more accurately; the hydrological changes in the basin can be reflected in real time to provide a basis for decision-making; and scenario analysis can be performed for different terrain and climate conditions to better understand water flow behavior.

[0121] S152. Combining the physical mechanism algorithm and the local runoff water fluid to simulate the scouring effect of floods on the river channel and embankments, a real erosion simulation effect on the river channel is formed. The twin scene dynamically updates the simulated erosion, damage, and overflowing effects of the river channel and embankments to obtain simulation results.

[0122] The physical mechanism algorithm is applied to local runoff water fluid to simulate the scouring effect of floods on river channels and embankments. This process will form a realistic erosion simulation effect on the river channel and dynamically update the twin scene to display the erosion, damage and overflow of the river channel and embankments.

[0123] The simulation results can reflect the changes in river channels and levees in real time, making it easier to monitor their status; help assess the potential risks of floods to infrastructure and provide important safety information; and provide decision support for water resource management and disaster response, as well as optimize river management and levee maintenance strategies.

[0124] Through these two steps, the final simulation results can not only effectively simulate the hydrological process in the basin, but also provide valuable data support for flood management and river maintenance, helping relevant decision makers to make scientific and reasonable plans and response measures.

[0125] Using the time-series characteristic water condition data calculated in real time by AI basin water conditions, the twin computable simulation engine is used to simulate the surface precipitation and water fluid in the basin. According to the physical characteristics of the scene simulation such as terrain, landforms, vegetation and soil, local runoff water fluids are formed. These water fluids flow from high to low relying on the terrain and landforms, and they continue to converge into small streams, gradually forming simulated water fluids with kinetic energy and potential energy. After convergence and acceleration, they finally converge into a rough riverbed. Between different riverbeds and embankments, water fluids realize the conversion of kinetic energy and potential energy through convergence and friction.

[0126] The twin computable engine can simulate various geological features, including topography, river networks and levees, as well as their associated physical parameters. These parameters include the geometry, material properties and boundary conditions of the levees, as well as different geological conditions within the watershed, such as soil structure and permeability, vegetation type and coverage, and riverbed characteristics with different substrates, such as gravel-filled riverbeds and bare riverbeds.

[0127] On this basis, a numerical model of coupled hydrodynamics and soil mechanics was established to describe the interaction between the flood flow field and the embankment. The hydrodynamic model uses shallow water equations to consider the unsteady flow characteristics of the flood, while the soil mechanics model uses the finite element method to deal with the nonlinear constitutive relationship of the embankment soil. By combining the physical simulation engine with the twin computable hydrodynamic algorithm, the confluence and scouring process of surface runoff and the physical action process of flood fluid in the river channel can be effectively simulated.

[0128] Specifically, AI technology is used to calculate the water flow data of the basin in real time, combined with hydrological data with time series characteristics, to develop an efficient twin computational simulation engine to simulate the precipitation and water flow dynamics in the basin. Through this engine, a detailed water flow model can be formed based on the actual physical characteristics of terrain, landforms, vegetation and soil, and its behavior in different environments can be analyzed.

[0129] The process of water fluid simulation is as follows:

[0130] Water flow behavior: Water flows from high to low, and in the process it constantly converges to form a stream. During the flow, the kinetic energy and potential energy of water are constantly converted, especially in a rough riverbed, where this conversion is particularly obvious.

[0131] Topography and environmental factors: The simulation process takes into account the geometry and physical properties of geological features, river networks, levees, etc. This includes soil structure, permeability, vegetation type and coverage, and the nature of the riverbed substrate (such as the difference between sand and gravel and bare).

[0132] Numerical model establishment: Establish a numerical model for coupling hydrodynamics and soil mechanics, including:

[0133] Hydrodynamic model: The shallow water equation is used to analyze the unsteady flow characteristics of floods.

[0134] Soil mechanics model: The finite element method is used to consider the nonlinear constitutive relationship of the embankment soil to more accurately describe the behavior of the embankment under flood action.

[0135] For embankment erosion simulation, the twin computational engine simulates the entire process of embankment erosion, involving multiple key algorithms:

[0136] Flood flow algorithm: Analyze the turbulent characteristics and dynamic changes of flood flow through shallow water equations.

[0137] Embankment seepage algorithm: Based on Darcy's law, the seepage process of flood through the embankment soil is calculated and the impact of water flow on the embankment is analyzed.

[0138] Embankment stability algorithm: Use the limit equilibrium method to evaluate the stability of the embankment under flood erosion and determine the risk of landslide.

[0139] Levee scour algorithm: Use empirical formulas to simulate flood scour on the levee surface and analyze the erosion process.

[0140] Levee failure algorithm: Apply the finite element method to evaluate the failure process of levees under severe scour and simulate potential dam failure phenomena.

[0141] During the flood process, the physical model is continuously updated by calculating the results of the above algorithms in real time. This enables the simulation to dynamically reflect the impact of the flood on the river channel and embankments, including:

[0142] The degree of erosion in the river channel; the damage to the levees; the effects of overtopping.

[0143] The main process is as follows:

[0144] The upstream precipitation process is calculated by combining the above-mentioned models to obtain the precipitation, infiltration and runoff of each network in the basin. Under the action of gravity and fluid, the precipitation is continuously gathered into the downstream river, forming the process of surface runoff, river flow and convergent flood.

[0145] The water flow process is combined with the soil erosion process to simulate the carrying of upstream sediment and other materials to the downstream river channel, forming a simulated flood fluid. In the twin computable simulation basin, the twin physical simulation engine is used to simulate physical phenomena in the real world in mathematical models and algorithms, mainly including rigid body dynamics, soft body dynamics and fluid dynamics. They can handle the effects of collisions, gravity, friction and other forces between objects, so that objects in the virtual environment can interact with each other in a real way. Through simulation calculation and simulation, a twin simulation scene effect of river network and embankment erosion and scour close to reality is formed.

[0146] Through physical simulation and calculation, we can simulate the actual erosion and damage of river channels and embankments. In the river network area where flood fluid flows, if the current river network section cannot carry the current amount of fluid, it will cause the embankment to overflow. Using the calculation of the twin engine and the analysis and perception of the AI ​​model, we can quickly form predictions and prejudgments of river disaster risk points.

[0147] This comprehensive simulation method can effectively predict and analyze the impact of floods on structures within the basin, providing a scientific basis for flood control design and management. The application of the twin computational physics simulation engine will provide strong support for real-time monitoring and emergency response of hydrological models.

[0148] S160. Identify risk points of the river in the basin according to the simulation results.

[0149] In this embodiment, the AI ​​model is used to analyze the river engineering working condition parameters in the simulation results to determine the danger level of the eroded river section and identify the risk points of the river in the basin.

[0150] Specifically, feature extraction: extract key operating parameters from the simulation results, such as precipitation, vegetation coverage, flow velocity, flow rate, soil type, soil moisture, etc.

[0151] Machine learning model training: Use historical data to train machine learning models, such as random forests, support vector machines, or neural networks, to identify erosion risks under different conditions. The model can learn different rainfall conditions, water conditions, watershed characteristics, river networks, and other identification features related to erosion levels to make predictions.

[0152] Risk level assessment: Through the trained model, the current working conditions are assessed and the erosion risk level of each river section is output. This is achieved by setting risk thresholds, dividing the risk level into three levels: low, medium, and high, to identify and predict risk points during simulation prediction.

[0153] By analyzing the working parameters of river engineering generated by AI, the danger level of eroded river sections can be predicted and evaluated. This timely risk point identification and early warning system can effectively identify the potential risks of river channels and embankments, thereby buying valuable prevention and risk elimination time for flood control and disaster reduction, thereby reducing the occurrence of flood disasters in the basin.

[0154] The method of this embodiment focuses on large-scale precipitation prediction and forecasting. By analyzing the relationship between precipitation and the defense water level and warning water level of rivers, lakes, reservoirs, and levees, the inundation range, water depth, and water receding time of floods and waterlogging are simulated. Combined with real-time monitoring data, the flood level is divided and the disaster loss is estimated to support rescue and disaster reduction decisions. Construct a twin computing scenario of all elements of a small watershed, use AI computing to train simulation models, dynamically simulate precipitation and surface runoff, and combine flood fluid simulation technology to accurately identify risk points in the watershed for real-time prediction and prejudgment. Through the twin computable engine, the real environment of the watershed is restored, and the precipitation, infiltration and convergence processes are calculated in real time to form a real flood fluid model, improve the intelligent identification and control efficiency of river flood prevention and disaster reduction, and ensure accurate response to flood risks. At the same time, flood simulation has physical properties, and simulates the entire process of the impact of floods on the surrounding environment based on dynamic and gravity characteristics.

[0155] Through the twin computable engine, the real terrain, landform, vegetation and river environment are constructed, and the precipitation, rainwater infiltration and surface convergence processes can be simulated in real time. The computing power of the AI ​​big model can be used to timely identify the flood risk of the river and issue early warning.

[0156] In addition, this method simulates the physical properties of floods and, based on the dynamic and gravity characteristics, truly reproduces the impact of floods on the surrounding environment, realizes the intelligent identification and management of the entire process of river flood control and disaster reduction, thereby improving the efficiency and accuracy of flood control and disaster reduction.

[0157] The above-mentioned AI identification method for river flood control and disaster reduction risk points based on the digital twin computable engine obtains the basic data of three-dimensional scenes and water conservancy project parameters in the basin, combines real-time rainfall, water conditions, soil moisture content and other monitoring data, establishes physical entities and twin simulation scenes based on these data, and calculates real-time water conditions through AI simulation to generate simulation results. Finally, the simulation results are used to identify risk points in the river within the basin to solve the problem that digital twin technology lacks intelligent risk identification and multi-factor integration in river flood control and disaster reduction, which affects the accuracy of prediction.

[0158] Figure 6 FIG. 1 is a flow chart of a method for identifying river flood control and disaster reduction risk points based on a digital twin computational engine according to another embodiment of the present invention. Figure 6 As shown, the AI ​​model identification method for river flood control and disaster reduction risk points based on the digital twin computable engine of this embodiment includes steps S210-S270. Steps S210-S260 are similar to steps S110-S160 in the above embodiment and are not repeated here. The added step S270 in this embodiment is described in detail below.

[0159] S270, combining the simulation results with the river risk points, and outputting a combined image.

[0160] Combining the simulation results with the risk points in the river channel can intuitively display potential risks, improve decision-making efficiency, and help relevant personnel quickly identify problems and take corresponding measures, thereby enhancing flood control safety.

[0161] The digital twin computational engine is a technology that combines physical entities, water conservancy projects and their digital models in a watershed with water conservancy flood control. Through real-time data collection and analysis, it can quickly simulate surface runoff and soil infiltration in the watershed and generate simulated flood fluids. In this engine scenario, combined with the topography, landforms, geology, vegetation and river channels of the watershed, flood fluids and underground infiltration data are gathered in real time to form a simulation of flood kinetic energy and potential energy from upstream to downstream, and analyze its flow and scouring on the surrounding terrain and its impact on wetlands, artificial lakes, reservoirs and rivers. With the help of AI computing power, the engine can timely and dynamically identify the risk points of river overflow, leakage and dam breach, thereby optimizing flood prevention and disaster reduction decisions. In addition, with the help of this twin computing engine, not only can the rainfall and water conditions of the watershed be simulated in real time, but also the rationality and safety of river engineering design can be evaluated by simulating different rainfall conditions, water conditions and water conservancy conditions, and the river network and water conservancy engineering design scheme can be optimized, thereby improving the efficiency and accuracy of flood prevention and disaster reduction.

[0162] Figure 7 is a schematic block diagram of an AI identification device 300 for river flood control and disaster reduction risk points based on a digital twin computable engine provided by an embodiment of the present invention. Figure 7 As shown, corresponding to the above-mentioned AI identification method for river flood control and disaster reduction risk points based on digital twin computable engine, the present invention also provides an AI identification device 300 for river flood control and disaster reduction risk points based on digital twin computable engine. The AI ​​identification device 300 for river flood control and disaster reduction risk points based on digital twin computable engine includes a unit for executing the above-mentioned AI identification method for river flood control and disaster reduction risk points based on digital twin computable engine, and the device can be configured in a server. Specifically, please refer to Figure 7 The river flood control and disaster reduction risk point AI identification device 300 based on the digital twin computable engine includes a basic data acquisition unit 301, a real-time data acquisition unit 302, an establishment unit 303, a simulation unit 304, a simulation unit 305 and an identification unit 306.

[0163] The basic data acquisition unit 301 is used to acquire the basic data of the three-dimensional scene and the water conservancy project parameters in the basin; the real-time data acquisition unit 302 is used to acquire the real-time rainfall, water conditions, soil moisture content, reservoir conditions, and levee monitoring data to obtain real-time data; the establishment unit 303 is used to establish the physical entity and the twin simulation scene according to the three-dimensional scene basic data and the water conservancy project parameters; the simulation unit 304 is used to perform AI simulation calculation of the water conditions according to the real-time data to obtain the simulation result; the simulation unit 305 is used to perform twin computable simulation according to the physical entity, the twin simulation scene and the simulation result to obtain the simulation result; the identification unit 306 is used to identify the risk points of the river in the basin according to the simulation result.

[0164] In one embodiment, if Figure 8 As shown, the establishment unit 303 includes a model construction subunit 3031 , an association subunit 3032 , a scene construction subunit 3033 and an integration subunit 3034 .

[0165] The model building subunit 3031 is used to build a physical entity model based on the three-dimensional scene basic data; the association subunit 3032 is used to associate the water conservancy project parameters with the physical entity model; the scene building subunit 3033 is used to build a twin simulation scene; the integration subunit 3034 is used to integrate the associated physical entity model into the twin simulation scene to obtain a physical entity and a twin simulation scene.

[0166] In one embodiment, if Fig. 9 As shown, the simulation unit 304 includes a first simulation subunit 3041 , a second simulation subunit 3042 , a prediction subunit 3043 and a combination subunit 3044 .

[0167] The first simulation subunit 3041 is used to use the CREST model to simulate the hydrological process of the basin, and through its distributed characteristics, integrate the runoff and runoff processes to generate hydrological results; the second simulation subunit 3042 is used to apply the SWAT model to quickly analyze soil moisture, runoff and erosion processes to obtain time series analysis results; the prediction subunit 3043 is used to use the CREST model to simulate the hydrology of the basin and the SWAT model to analyze the result data of soil moisture, runoff and erosion processes, use machine learning algorithms to predict basin runoff and surface runoff, and simultaneously tune the model parameters to improve the accuracy of basin runoff and surface erosion; the combination subunit 3044 is used to combine the hydrological process results, the time series runoff analysis results, and the simulation and prediction of the model correction, and finally form precipitation runoff and runoff simulation results that are consistent with the basin.

[0168] In one embodiment, if Fig.10As shown, the simulation unit 305 includes a fluid simulation subunit 3051 and an erosion simulation subunit 3052 .

[0169] The fluid simulation subunit 3051 is used to simulate the surface precipitation and water fluid in the basin according to the simulation results by using the twin computable simulation engine, and to form a local runoff water fluid according to the physical entity and the twin simulation scene; the erosion simulation subunit 3052 is used to combine the physical mechanism algorithm and the local runoff water fluid to simulate the scouring effect of floods on the river channel and embankment, thereby forming a real erosion simulation effect on the river channel. The twin scene dynamically updates the simulated erosion, damage, and overflowing effects of the river channel and embankment to obtain the simulation results.

[0170] In one embodiment, the identification unit 306 is used to analyze the river engineering working condition parameters in the simulation results using an AI model to determine the danger level of the eroded river section and identify risk points in the river within the basin.

[0171] Fig.11 is a schematic block diagram of an AI model identification device for river flood control and disaster reduction risk points based on a digital twin computable engine provided by another embodiment of the present invention. Fig.11 As shown, the river flood control and disaster reduction risk point AI model identification device based on the digital twin computable engine of this embodiment adds an output unit 307 on the basis of the above embodiment.

[0172] The output unit 307 is used to combine the simulation result with the river risk point and output the combined picture.

[0173] It should be noted that technical personnel in the relevant field can clearly understand that the specific implementation process of the above-mentioned river flood control and disaster reduction risk point AI identification device 300 based on the digital twin computable engine and each unit can refer to the corresponding description in the aforementioned method embodiment, and for the convenience and conciseness of the description, it will not be repeated here.

[0174] The above-mentioned river flood control and disaster reduction risk point AI identification device 300 based on digital twin computable engine can be implemented in the form of a computer program, which can be used in Fig.12 Runs on the computer device shown.

[0175] See also Fig.12 , Fig.12 5 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device 500 may be a server, wherein the server may be an independent server or a server cluster composed of multiple servers.

[0176] See also Fig.12The computer device 500 includes a processor 502 , a memory and a network interface 505 connected via a system bus 501 , wherein the memory may include a non-volatile storage medium 503 and an internal memory 504 .

[0177] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions, which, when executed, can enable the processor 502 to execute an AI identification method for river flood control and disaster reduction risk points based on a digital twin computable engine.

[0178] The processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500 .

[0179] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute an AI identification method for river flood control and disaster reduction risk points based on a digital twin computable engine.

[0180] The network interface 505 is used to communicate with other devices over the network. Fig.12 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present application, and does not constitute a limitation on the computer device 500 to which the solution of the present application is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0181] The processor 502 is used to run the computer program 5032 stored in the memory to implement the following steps:

[0182] Obtain basic three-dimensional scene data and water conservancy project parameters in the basin; obtain real-time rainfall conditions, water conditions, soil moisture content, reservoir conditions, and levee monitoring data to obtain real-time data; establish physical entities and twin simulation scenes based on the three-dimensional scene basic data and water conservancy project parameters; perform AI simulation calculations of water conditions based on the real-time data to obtain simulation results; perform twin computable simulation based on the physical entities, twin simulation scenes and the simulation results to obtain simulation results; identify risk points in the river channel in the basin based on the simulation results.

[0183] Among them, the basic data of the three-dimensional scene includes river networks, lakes, wetlands, flood discharge areas, topography, geology, soil, vegetation types, buildings and levee information; the water conservancy project parameters include flood discharge area capacity, levee parameters, regional soil characteristics and historical rainfall and flood data.

[0184] In one embodiment, when the processor 502 implements the step of establishing the physical entity and the twin simulation scene according to the three-dimensional scene basic data and the water conservancy project parameters, it specifically implements the following steps:

[0185] Construct a physical entity model based on the three-dimensional scene basic data; associate the water conservancy project parameters with the physical entity model; construct a twin simulation scene; integrate the associated physical entity model into the twin simulation scene to obtain a physical entity and a twin simulation scene.

[0186] In one embodiment, when the processor 502 implements the step of performing AI simulation calculation of the water condition according to the real-time data to obtain the simulation result, the processor 502 specifically implements the following steps:

[0187] The CREST model is used to simulate the hydrological process of the basin. Through its distributed characteristics, the flow generation and runoff processes are integrated to generate hydrological results. The SWAT model is used to quickly analyze soil moisture, runoff and erosion processes to obtain time series analysis results. The CREST model is used to simulate the hydrology of the basin and the SWAT model is used to analyze the result data of soil moisture, runoff and erosion processes. The basin runoff and surface runoff are predicted using machine learning algorithms, and the model parameters are tuned to improve the accuracy of basin runoff and surface erosion. The hydrological process results, the time series runoff analysis results, and the simulation and prediction of the model correction are combined to finally form precipitation runoff and runoff simulation results that are consistent with the basin.

[0188] In one embodiment, when the processor 502 implements the step of performing twin computable simulation according to the physical entity, the twin simulation scenario, and the simulation result to obtain the simulation result, the processor 502 specifically implements the following steps:

[0189] According to the simulation results, the twin computable simulation engine is used to simulate the surface precipitation and water fluid in the basin, and based on the physical entity and the twin simulation scene, a local runoff water fluid is formed; the physical mechanism algorithm and the local runoff water fluid are combined to simulate the scouring effect of floods on rivers and embankments, forming a real erosion simulation effect on the river channel, and the twin scene is dynamically updated to simulate the erosion, damage, and overflowing effects of rivers and embankments to obtain simulation results.

[0190] In one embodiment, when the processor 502 implements the step of identifying risk points of a river channel in a watershed according to the simulation results, the processor 502 specifically implements the following steps:

[0191] The AI ​​model is used to analyze the river engineering operating parameters in the simulation results to determine the danger level of the eroded river section and identify the risk points of the river in the basin.

[0192] In one embodiment, after implementing the step of identifying risk points of the river in the basin according to the simulation results, the processor 502 further implements the following steps:

[0193] The simulation results are combined with the river risk points, and a combined image is output.

[0194] It should be understood that in the embodiment of the present application, the processor 502 may be a central processing unit (CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0195] It can be understood by those skilled in the art that all or part of the processes in the method for implementing the above embodiment can be completed by instructing the relevant hardware through a computer program. The computer program includes program instructions, and the computer program can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiment of the above method.

[0196] Therefore, the present invention also provides a storage medium. The storage medium may be a computer-readable storage medium. The storage medium stores a computer program, wherein when the computer program is executed by a processor, the processor executes the following steps:

[0197] Obtain basic three-dimensional scene data and water conservancy project parameters in the basin; obtain real-time rainfall conditions, water conditions, soil moisture content, reservoir conditions, and levee monitoring data to obtain real-time data; establish physical entities and twin simulation scenes based on the three-dimensional scene basic data and water conservancy project parameters; perform AI simulation calculations of water conditions based on the real-time data to obtain simulation results; perform twin computable simulation based on the physical entities, twin simulation scenes and the simulation results to obtain simulation results; identify risk points in the river channel in the basin based on the simulation results.

[0198] Among them, the basic data of the three-dimensional scene includes river networks, lakes, wetlands, flood discharge areas, topography, geology, soil, vegetation types, buildings and levee information; the water conservancy project parameters include flood discharge area capacity, levee parameters, regional soil characteristics and historical rainfall and flood data.

[0199] In one embodiment, when the processor executes the computer program to implement the step of establishing a physical entity and a twin simulation scene according to the three-dimensional scene basic data and the water conservancy project parameters, the processor specifically implements the following steps:

[0200] Construct a physical entity model based on the three-dimensional scene basic data; associate the water conservancy project parameters with the physical entity model; construct a twin simulation scene; integrate the associated physical entity model into the twin simulation scene to obtain a physical entity and a twin simulation scene.

[0201] In one embodiment, when the processor executes the computer program to implement the step of performing AI simulation calculation of the water condition according to the real-time data to obtain the simulation result, the processor specifically implements the following steps:

[0202] The CREST model is used to simulate the hydrological process of the basin. Through its distributed characteristics, the flow generation and runoff processes are integrated to generate hydrological results. The SWAT model is used to quickly analyze soil moisture, runoff and erosion processes to obtain time series analysis results. The CREST model is used to simulate the hydrology of the basin and the SWAT model is used to analyze the result data of soil moisture, runoff and erosion processes. The basin runoff and surface runoff are predicted using machine learning algorithms, and the model parameters are tuned to improve the accuracy of basin runoff and surface erosion. The hydrological process results, the time series runoff analysis results, and the simulation and prediction of the model correction are combined to finally form precipitation runoff and runoff simulation results that are consistent with the basin.

[0203] In one embodiment, when the processor executes the computer program to implement the step of performing twin computable simulation according to the physical entity, the twin simulation scenario, and the simulation result to obtain the simulation result, the following steps are specifically implemented:

[0204] According to the simulation results, the twin computable simulation engine is used to simulate the surface precipitation and water fluid in the basin, and based on the physical entity and the twin simulation scene, a local runoff water fluid is formed; the physical mechanism algorithm and the local runoff water fluid are combined to simulate the scouring effect of floods on rivers and embankments, forming a real erosion simulation effect on the river channel, and the twin scene is dynamically updated to simulate the erosion, damage, and overflowing effects of rivers and embankments to obtain simulation results.

[0205] In one embodiment, when the processor executes the computer program to implement the step of identifying risk points of a river channel in a watershed according to the simulation results, the processor specifically implements the following steps:

[0206] The AI ​​model is used to analyze the river engineering operating parameters in the simulation results to determine the danger level of the eroded river section and identify the risk points of the river in the basin.

[0207] In one embodiment, after executing the computer program to implement the step of identifying risk points of a river channel in a watershed according to the simulation results, the processor further implements the following steps:

[0208] The simulation results are combined with the river risk points, and a combined image is output.

[0209] The storage medium may be a USB flash drive, a mobile hard disk, a read-only memory (ROM), a magnetic disk, or an optical disk, etc., which are computer-readable storage media that can store program codes.

[0210] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0211] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of each unit is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0212] The steps in the method of the embodiment of the present invention can be adjusted in order, combined and deleted according to actual needs. The units in the device of the embodiment of the present invention can be combined, divided and deleted according to actual needs. In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0213] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, terminal, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention.

[0214] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. An AI identification method for river flood control and disaster reduction risk points based on a digital twin computable engine, characterized in that: include: Obtain basic three-dimensional scene data and water conservancy project parameters within the basin; Obtain real-time rainfall, water conditions, soil moisture, reservoir conditions, and levee monitoring data to obtain real-time data; Establishing physical entities and twin simulation scenes according to the three-dimensional scene basic data and water conservancy project parameters; Perform AI simulation calculation of water conditions according to the real-time data to obtain simulation results; Performing twin computable simulation according to the physical entity, the twin simulation scenario and the simulation result to obtain a simulation result; Identify risk points of the river in the basin according to the simulation results; The AI ​​simulation calculation of the water situation is performed according to the real-time data to obtain the simulation result, including: The CREST model is used to simulate the hydrological process of the basin. Through its distributed characteristics, it integrates the flow generation and runoff processes to generate hydrological results. The SWAT model is used to quickly analyze soil moisture, runoff and erosion processes to obtain time series analysis results; Use the CREST model to simulate the hydrology of the watershed and the SWAT model to analyze the result data of soil moisture, runoff and erosion processes, use machine learning algorithms to predict watershed runoff and surface runoff, and optimize model parameters to improve the accuracy of watershed runoff and surface erosion; Combining the hydrological results, the analysis results of the time series, and the simulation and prediction of the model modification, ultimately forming the simulation results of precipitation flow and runoff that are consistent with the basin; The performing twin computable simulation according to the physical entity, the twin simulation scenario and the simulation result to obtain a simulation result includes: According to the simulation results, using a twin computable simulation engine to simulate surface precipitation and water fluid in the basin, and forming local runoff water fluid according to the physical entity and the twin simulation scenario; The physical mechanism algorithm and the local runoff water fluid are combined to simulate the scouring effect of floods on the river channel and embankments, forming a real erosion simulation effect on the river channel. The twin scene dynamically updates the simulated erosion, damage, and overflowing effects of the river channel and embankments to obtain simulation results.

2. The AI ​​identification method for river flood control and disaster reduction risk points based on a digital twin computable engine according to claim 1 is characterized in that: The basic data of the three-dimensional scene include river networks, lakes, wetlands, flood discharge areas, topography, geology, soil, vegetation types, buildings and levee information; the water conservancy project parameters include flood discharge area capacity, levee parameters, regional soil characteristics and historical rainfall and flood data.

3. The AI ​​identification method for river flood control and disaster reduction risk points based on a digital twin computable engine according to claim 1 is characterized in that: The establishing of physical entities and twin simulation scenes according to the three-dimensional scene basic data and water conservancy project parameters includes: Constructing a physical entity model according to the three-dimensional scene basic data; Associating the hydraulic engineering parameters with the physical entity model; Build twin simulation scenarios; The associated physical entity model is integrated into the twin simulation scene to obtain the physical entity and the twin simulation scene.

4. The AI ​​identification method for river flood control and disaster reduction risk points based on a digital twin computable engine according to claim 1 is characterized in that: The step of identifying risk points of a river in a watershed according to the simulation results includes: The AI ​​model is used to analyze the river engineering operating parameters in the simulation results to determine the danger level of the eroded river section and identify the risk points of the river in the basin.

5. The AI ​​identification method for river flood control and disaster reduction risk points based on a digital twin computable engine according to claim 1 is characterized in that: After identifying the risk points of the river in the basin according to the simulation results, the method further includes: The simulation results are combined with the river risk points, and a combined image is output.

6. An AI identification device for river flood control and disaster reduction risk points based on a digital twin computable engine, characterized in that: include: Basic data acquisition unit, used to obtain basic three-dimensional scene data and water conservancy project parameters in the basin; A real-time data acquisition unit is used to obtain real-time rainfall, water conditions, soil moisture content, reservoir conditions, and levee monitoring data to obtain real-time data; An establishing unit, used to establish a physical entity and a twin simulation scene according to the three-dimensional scene basic data and water conservancy project parameters; A simulation unit, used for performing AI simulation calculation of water conditions according to the real-time data to obtain simulation results; A simulation unit, configured to perform a twin computable simulation according to the physical entity, the twin simulation scenario and the simulation result to obtain a simulation result; An identification unit, used for identifying risk points of a river in a watershed according to the simulation results; The AI ​​simulation calculation of the water situation is performed according to the real-time data to obtain the simulation result, including: The CREST model is used to simulate the hydrological process of the basin. Through its distributed characteristics, it integrates the flow generation and runoff processes to generate hydrological results. The SWAT model is used to quickly analyze soil moisture, runoff and erosion processes to obtain time series analysis results; Use the CREST model to simulate the hydrology of the watershed and the SWAT model to analyze the result data of soil moisture, runoff and erosion processes, use machine learning algorithms to predict watershed runoff and surface runoff, and optimize model parameters to improve the accuracy of watershed runoff and surface erosion; Combining the hydrological results, the analysis results of the time series, and the simulation and prediction of the model modification, ultimately forming the simulation results of precipitation flow and runoff that are consistent with the basin; The performing twin computable simulation according to the physical entity, the twin simulation scenario and the simulation result to obtain a simulation result includes: According to the simulation results, using a twin computable simulation engine to simulate surface precipitation and water fluid in the basin, and forming local runoff water fluid according to the physical entity and the twin simulation scenario; The physical mechanism algorithm and the local runoff water fluid are combined to simulate the scouring effect of floods on the river channel and embankments, forming a real erosion simulation effect on the river channel. The twin scene dynamically updates the simulated erosion, damage, and overflowing effects of the river channel and embankments to obtain simulation results.

7. A computer device, characterized in that: The computer device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 5 when executing the computer program.

8. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images