Intelligent management system for water conservancy projects

Abstract

The invention discloses an intelligent management system for water conservancy engineering projects, which relates to the technical field of water conservancy engineering project management. Through a plurality of modules such as a data monitoring module, a three-dimensional virtual simulation construction module, a sedimentation monitoring module, and a groundwater infiltration simulation module, the operation of a reservoir under different rainfall and water level conditions is comprehensively analyzed, and the stress performance of a reservoir dam body is monitored and predicted in real time. The simulation accuracy evaluation module determines the simulation accuracy according to the influence of sedimentation and groundwater infiltration, and the simulation result classification module classifies and manages the simulation results. Through the management module, the system formulates reservoir scheduling and decision-making based on accurate simulation results, and for incompletely accurate simulation results, the dynamic adjustment module analyzes the simulation deviation and adjusts the water storage capacity and the bearing capacity of the dam body, thereby optimizing the design and construction, reducing rework and resource waste caused by improper design, and improving the safety and operation efficiency of the reservoir.

Description

Technical Field

[0001] The present invention relates to the technical field of water conservancy project management, and in particular to an intelligent management system for water conservancy project. Background Art

[0002] Intelligent management of water conservancy projects refers to the use of advanced intelligent technologies, such as the Internet of Things, big data, artificial intelligence and cloud computing, to improve the management efficiency of various links such as planning, design, construction, operation and maintenance of water conservancy projects. In the existing technology, during the construction of water conservancy projects, by using virtual simulation technology, a digital model of a water conservancy project can be created in a computer, and its future operation process and various possible scenarios can be simulated in a virtual environment. Through this simulation, engineers can predict the performance of the project, potential problems and its impact on the surrounding environment in advance, and then optimize and adjust the design and construction plan to avoid mistakes and unnecessary costs in actual operations.

[0003] Specifically, in the design of reservoirs, virtual simulation can simulate the water storage and drainage performance of reservoirs under different rainfall conditions, predict the stress conditions of the dam, and thus provide a basis for optimizing the dam design. In this way, rework or waste of resources caused by unreasonable design during construction can be avoided. However, the calculation of dam stress is very complicated, including water pressure, earthquake impact, soil pressure, etc. If the virtual simulation does not accurately simulate the stress conditions of the reservoir dam under different operating conditions, the actual pressure on the dam may be underestimated, resulting in cracking, settlement, and even structural failure of the dam in long-term operation. At the same time, the incorrect estimation of the stress on the dam by virtual simulation may affect the water storage scheduling of the reservoir. If the simulation fails to correctly calculate the bearing capacity of the dam, the reservoir may be forced to limit the water storage to avoid safety risks. This will result in the inability to fully utilize the regulation and storage function of the reservoir and affect the effective management of water resources. Summary of the invention

[0004] The purpose of the present invention is to provide an intelligent management system for water conservancy projects to solve the shortcomings of the background technology.

[0005] In order to achieve the above-mentioned object, the present invention provides the following technical solutions: an intelligent management system for water conservancy projects, comprising a data monitoring module, a three-dimensional virtual simulation construction module, a sediment deposition monitoring module, a groundwater infiltration simulation module, a simulation accuracy evaluation module, a simulation result division module, a management module and a dynamic adjustment module;

[0006] Data monitoring module: divide the water conservancy project area into several sub-project areas, monitor and obtain the three-dimensional terrain data and hydrological and meteorological data of each sub-project area in different time periods;

[0007] 3D virtual simulation construction module: Build a 3D virtual simulation model, set different rainfall scenarios based on historical rainfall data and future weather forecasts, and input different rainfall scenarios into the hydrodynamic model to simulate reservoir storage and drainage behavior;

[0008] Sedimentation monitoring module: In the 3D virtual simulation process, the sedimentation rate in the reservoir is monitored under different rainfall and flow conditions, the abnormal change trend of sedimentation rate is determined, and the interference degree of sedimentation on the stability of the dam is evaluated;

[0009] Groundwater seepage simulation module: Use the groundwater seepage simulation tool to simulate the groundwater flow around the dam body, analyze the fluctuation amplitude of groundwater seepage pressure under different water levels and rainfall conditions, predict the risk of dam foundation sliding under different groundwater levels, and evaluate the impact of groundwater seepage on the force distribution of the dam body;

[0010] Simulation accuracy evaluation module: Comprehensively analyze the interference degree of sediment deposition on the stability of the dam body and the influence of groundwater infiltration on the stress distribution of the dam body, and determine the simulation accuracy of the virtual simulation of the stress conditions of the reservoir dam body under different operating conditions based on the analysis results;

[0011] Simulation result division module: divides the simulation accuracy of the virtual simulation on the stress conditions of the reservoir dam under different operating conditions into accurate simulation results, incompletely accurate simulation results and inaccurate simulation results, and manages them accordingly;

[0012] Management module: For accurate simulation results, the management module directly makes decisions and scheduling strategies based on the simulation results; for inaccurate simulation results, a warning should be triggered immediately and the simulation model should be recalibrated;

[0013] Dynamic adjustment module: For inaccurate simulation results, the simulation accuracy of the virtual simulation of the stress conditions of the reservoir dam under different operating conditions within a fixed time period is deeply analyzed, and the abnormal degree of simulation deviation is judged according to the analysis results. If the abnormal degree is high, the water storage capacity and dam bearing capacity are dynamically adjusted.

[0014] Preferably, in the sediment deposition monitoring module, the sediment deposition rate of each area is compared with the data under different flow rates and different water depths of the reservoir to generate a sediment deposition rate anomaly index. The method for obtaining the sediment deposition rate anomaly index is:

[0015] The sediment deposition rate data of different areas of the reservoir are obtained through simulation monitoring. The sediment deposition rate data is time series data, which indicates the change of sediment deposition rate in different periods. The initial data set is selected and the average sediment deposition rate of the initial data set is calculated. , expressed as: ;in, is the sedimentation rate in the ith period, n is the total sample size, and the standard deviation of the sedimentation rate is calculated. , the formula is: ; Determine the control limits of the control chart. The control chart has three control lines, namely the center line CL, the upper control limit UCL and the lower control limit LCL. The center line is the average value of the sedimentation rate. The upper control limit is set to the average value plus 3 times the standard deviation; the lower control limit is set to the average value minus 3 times the standard deviation. Draw a control chart with the X-axis as the time series and the Y-axis as the sediment deposition rate for each period. The sediment deposition rate at each time point is plotted on the chart. When the sediment deposition rate data point When it is greater than the upper control limit UCL or less than the lower control limit LCL, it is considered an abnormal point; the sediment deposition rate anomaly index is marked as HK, and the sediment deposition rate data point When it is greater than the upper control limit UCL, ; When the sediment deposition rate data point When it is less than the lower control limit LCL, .

[0016] Preferably, in the groundwater infiltration simulation module, the infiltration pressure data under normal water level and extreme water level, normal rainfall and extreme rainfall conditions are obtained, and the infiltration pressure fluctuation amplitude of groundwater under different water level and rainfall conditions is analyzed to generate the infiltration pressure fluctuation index. The method for obtaining the infiltration pressure fluctuation index is:

[0017] Obtain osmotic pressure data within a time period of W , including pressure values ​​under different water levels and rainfall conditions, and obtaining the prior distribution of seepage pressure based on historical data, expressed as the mean of the pressure and variance Based on the groundwater seepage model, the distribution form of the seepage pressure is defined, and the seepage pressure obeys the normal distribution ; Prior distribution represents the initial understanding of osmotic pressure when there is no new observation data, constructing the likelihood function, the likelihood function Describes the possibility of observing the current osmotic pressure data under given parameter conditions. The expression is: ; is the ith observed osmotic pressure data point, and m is the number of observed data. Apply Bayes’ theorem to calculate the posterior distribution. Updated by prior distribution and likelihood function, the expression is: ; is the prior distribution of groundwater pressure, is the marginal likelihood of the observed data; update the parameters and obtain the posterior distribution through Bayes’ theorem, the expressions are: ; ; In the formula, is the updated mean osmotic pressure, is the updated osmotic pressure variance, is the variance of the observed data, and the osmotic pressure fluctuation index is calculated as follows: ; Where GH is the osmotic pressure fluctuation index.

[0018] Preferably, in the simulation accuracy evaluation module, the sediment deposition rate anomaly index and the seepage pressure fluctuation index are converted into a first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses each group of first eigenvectors to predict the simulation accuracy value label of the virtual simulation of the stress conditions of the reservoir dam body under different operating conditions as the prediction target, and takes minimizing the sum of the prediction errors of the simulation accuracy value labels of the stress conditions of the reservoir dam body under different operating conditions for all virtual simulations as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The simulation accuracy value of the virtual simulation of the stress conditions of the reservoir dam body under different operating conditions is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0019] Preferably, in the simulation result division module, the simulation accuracy of the virtual simulation on the stress condition of the reservoir dam body under different operating conditions is divided into accurate simulation results, incompletely accurate simulation results and inaccurate simulation results, specifically:

[0020] The obtained simulation accuracy value of the virtual simulation on the stress condition of the reservoir dam body under different operating conditions is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the simulation accuracy value of the virtual simulation on the stress condition of the reservoir dam body under different operating conditions is compared with the first standard threshold and the second standard threshold respectively;

[0021] If the simulation accuracy value of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is greater than the second standard threshold value, it means that the simulation accuracy of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is high, and a high-accuracy model signal is generated at this time, and the simulation results of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions are classified as accurate simulation results;

[0022] If the simulation accuracy value of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the simulation accuracy of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is general, and a medium accuracy model signal is generated at this time, and the simulation results of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions are classified as incompletely accurate simulation results;

[0023] If the simulation accuracy value of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is less than the first standard threshold, it means that the simulation accuracy of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is low. At this time, a low-accuracy model signal is generated, and the simulation results of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions are classified as inaccurate simulation results.

[0024] Preferably, in the dynamic adjustment module, for inaccurate simulation results, the simulation accuracy of the virtual simulation of the stress conditions of the reservoir dam under different operating conditions within a fixed time period is deeply analyzed, and the abnormal degree of simulation deviation is judged according to the analysis results, specifically:

[0025] For inaccurate simulation results, set the simulation prediction value To simulate the predicted value of the dam body stress condition within time t, the actual monitoring value The actual monitoring results show the stress of the dam body; the simulation deviation represents the difference between simulation prediction and actual monitoring and is defined as: ; Calculate the deviation rate, the expression is: ; is the deviation rate, which indicates the percentage deviation of the simulation result from the actual monitoring value; the deviation rate in the subsequent fixed time period T is accumulated and summed to calculate the cumulative deviation in the T time period , the expression is: ; Calculate the abnormality index based on the cumulative deviation and deviation rate , the expression is: ; In the formula, is the average value of the actual force value in the time period T.

[0026] Preferably, if the abnormal index , it is necessary to dynamically adjust the reservoir water storage capacity, is the preset abnormal index reference threshold, and the current water storage capacity is set to , the maximum safe water storage capacity is , according to the degree of abnormality, the formula for dynamically adjusting the water storage capacity is: ; where α is the adjustment coefficient, is the adjusted water storage capacity;

[0027] Adjust the dam bearing capacity based on the anomaly index , assuming the current carrying capacity is , the formula for adjusting the carrying capacity under abnormal conditions is: ; where β is the load-bearing capacity adjustment factor, is the adjusted bearing capacity of the dam;

[0028] For the safe operation range of the dam body, the upper limits of water storage and carrying capacity are redefined through simulation adjustment. If the anomaly persists, the upper limit of water storage will be further reduced. and carrying capacity limit , the expressions are: ; ; where γ and δ are control coefficients for long-term adjustment, For maximum safe water storage capacity, is the maximum bearing capacity of the dam.

[0029] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0030] 1. The present invention obtains three-dimensional terrain and hydrological and meteorological data through the data monitoring module, combines the three-dimensional virtual simulation construction module to simulate the water storage and drainage performance under different rainfall conditions, and the sediment deposition monitoring module and groundwater infiltration simulation module to monitor the stability and stress of the dam body in real time, and comprehensively predict the performance of the reservoir dam body under different operating conditions. The simulation accuracy evaluation module further ensures the reliability of the simulation model by analyzing the influence of sediment deposition and seepage pressure, and manages the classification and application of simulation results through the simulation result classification module.

[0031] 2. On the one hand, based on the accuracy of the simulation results, the management module can optimize the reservoir's water storage scheduling and dam carrying capacity in real time to avoid resource waste and safety hazards caused by design defects; on the other hand, the system can respond to the dynamic changes of the reservoir according to the simulation error. When the deviation exceeds the threshold, the water storage and carrying capacity are adjusted in time to ensure the long-term safe operation of the reservoir. Through automatic calibration and dynamic adjustment, the system improves the management efficiency of the reservoir, reduces rework costs, and effectively enhances the management and utilization capabilities of water resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0033] Figure 1It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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.

[0035] For examples, see Figure 1 As shown, the intelligent management system for water conservancy projects described in this embodiment includes a data monitoring module, a three-dimensional virtual simulation construction module, a sediment deposition monitoring module, a groundwater infiltration simulation module, a simulation accuracy evaluation module, a simulation result division module, a management module and a dynamic adjustment module;

[0036] Data monitoring module: divide the water conservancy project area into several sub-project areas, monitor and obtain the three-dimensional terrain data and hydrological and meteorological data of each sub-project area in different time periods;

[0037] 3D virtual simulation construction module: Build a 3D virtual simulation model, set different rainfall scenarios based on historical rainfall data and future weather forecasts, and input different rainfall scenarios into the hydrodynamic model to simulate reservoir storage and drainage behavior;

[0038] Sedimentation monitoring module: In the 3D virtual simulation process, the sedimentation rate in the reservoir is monitored under different rainfall and flow conditions, the abnormal change trend of sedimentation rate is determined, and the interference degree of sedimentation on the stability of the dam is evaluated;

[0039] Groundwater seepage simulation module: Use the groundwater seepage simulation tool to simulate the groundwater flow around the dam body, analyze the fluctuation amplitude of groundwater seepage pressure under different water levels and rainfall conditions, predict the risk of dam foundation sliding under different groundwater levels, and evaluate the impact of groundwater seepage on the force distribution of the dam body;

[0040] Simulation accuracy evaluation module: Comprehensively analyze the interference degree of sediment deposition on the stability of the dam body and the influence of groundwater infiltration on the stress distribution of the dam body, and determine the simulation accuracy of the virtual simulation of the stress conditions of the reservoir dam body under different operating conditions based on the analysis results;

[0041] Simulation result division module: divides the simulation accuracy of the virtual simulation on the stress conditions of the reservoir dam under different operating conditions into accurate simulation results, incompletely accurate simulation results and inaccurate simulation results, and manages them accordingly;

[0042] Management module: For accurate simulation results, the management module directly makes decisions and scheduling strategies based on the simulation results; for inaccurate simulation results, a warning should be triggered immediately and the simulation model should be recalibrated;

[0043] Dynamic adjustment module: For inaccurate simulation results, the simulation accuracy of the virtual simulation of the stress conditions of the reservoir dam under different operating conditions within a fixed time period is deeply analyzed, and the abnormal degree of simulation deviation is judged according to the analysis results. If the abnormal degree is high, the water storage capacity and dam bearing capacity are dynamically adjusted.

[0044] In the data monitoring module, the purpose of dividing the area is clearly defined according to the project objectives of the water conservancy project. Different regional divisions may be based on objectives such as topography, hydrological conditions, construction difficulty, and management efficiency. Generally, the purpose of dividing the area is to more finely manage the project implementation, risk control, and data monitoring of each area. Obtain preliminary data for the area, including basic data such as the total area of ​​the project, topographic maps, geological survey reports, hydrological and meteorological data, and watershed division maps. You can also refer to satellite images and aerial images to understand the natural conditions of the overall area.

[0045] Determine the classification criteria, including:

[0046] Topographic standards: The area is divided according to its topographic features, such as mountains, hills, plains, river valleys, etc. Areas with complex topography can be divided into smaller sub-project areas for more refined management.

[0047] Hydrological condition standards: Consider hydrological characteristics such as watershed division, river channel, and catchment area. Different areas within a watershed may have different water flow directions and water storage capacities, and can be divided according to the hydrodynamic characteristics of the watershed.

[0048] Functional division standard: The project is divided according to different functional areas, such as reservoir dam site, water storage area, spillway, drainage facilities, ecological protection area, etc. Each functional area may involve different management requirements and technical priorities.

[0049] Construction management standards: Consider the feasibility of engineering construction and divide the sub-project area according to the construction zone. This is usually based on factors such as the entry and exit routes of construction equipment and the division of construction stages (such as initial engineering, mid-term engineering, and final completion).

[0050] Import the collected basic data such as topography, hydrology, and watershed into GIS (Geographic Information System) software, such as ArcGIS, QGIS, etc. In the GIS system, set the division standards (such as terrain height, watershed boundaries, and functional zoning), and use geographic analysis tools to automatically or manually divide. Based on the set standards, generate boundary maps of subproject areas, and mark the functions and management requirements of different areas in GIS. Number each subproject area and set its specific attributes, such as area, altitude, drainage direction, etc., for subsequent management and data collection.

[0051] Monitor and obtain three-dimensional terrain data and hydrological and meteorological data in the subproject area. The data acquisition methods include:

[0052] Laser Radar (LiDAR) mapping: Use airborne or drone-mounted LiDAR equipment to conduct high-precision three-dimensional terrain mapping of the subproject area.

[0053] The LiDAR system captures elevation data of the terrain surface by emitting laser pulses and generates a high-precision digital elevation model (DEM), which usually has an accuracy of centimeters.

[0054] In mountainous areas or areas with heavy vegetation, LiDAR can penetrate the vegetation and obtain detailed elevation information of the surface.

[0055] Drone aerial mapping: Use drones equipped with high-definition cameras and photogrammetry technology to generate high-resolution images of the area through aerial photography. Through multi-angle aerial photography, a three-dimensional terrain model of the area can be generated. The images are then processed through software (such as Pix4D, Agisoft Metashape) to generate a three-dimensional point cloud and digital elevation model.

[0056] Satellite remote sensing data: Use high-resolution satellite image data, combined with GIS tools, to obtain large-scale three-dimensional terrain data. Satellite images such as Landsat and Sentinel provide regional elevation information, but the accuracy is low and is suitable for large-scale preliminary mapping.

[0057] For areas with higher precision requirements, LiDAR or drone aerial photography can be used as a supplement.

[0058] After data collection is completed, 3D modeling software (such as Bentley ContextCapture, AutoCAD Civil 3D) is used to process the collected data. Data from different sources (LiDAR, drones, satellite data) are integrated to generate a unified 3D terrain model covering the sub-project area of ​​the entire water conservancy project. Information such as terrain elevation, slope, and surface features are added to the model for subsequent analysis and simulation.

[0059] Meteorological data acquisition:

[0060] Automatic weather station: An automatic weather station is installed in each subproject area or its surroundings to collect real-time data such as temperature, precipitation, wind speed, humidity, etc. The automatic weather station can transmit data to the central management system through wireless communication.

[0061] Satellite meteorological data: Obtain regional atmospheric monitoring data through meteorological satellites (such as NOAA, Meteosat), including rainfall distribution, temperature changes, cloud movement, etc. These data can be used for the analysis of historical meteorological trends and the prediction of future weather.

[0062] Meteorological data integration: Combine data from automatic weather stations with satellite meteorological data to generate a refined weather forecast model, providing a reference for short-term weather forecasts and long-term climate trends.

[0063] Hydrological data acquisition:

[0064] Hydrological monitoring stations: Hydrological monitoring stations are installed near the main rivers, water storage areas and dam sites in the subproject area to monitor water level, flow, rainfall runoff, evaporation and other information in real time. Hydrological monitoring stations measure water level and flow rate changes through radar, ultrasonic or float equipment.

[0065] Remote sensing and watershed models: Remote sensing technology is combined with hydrological models (such as SWAT, HEC-HMS) to simulate the hydrological conditions of the entire basin, especially to analyze the rainfall-runoff process, reservoir water storage and flood discharge behavior.

[0066] Collect historical hydrological data from the past several years or even decades in the region, such as river flow changes, rainfall intensity, reservoir water storage, etc., as the basis for future trend prediction. Integrate and store the acquired hydrological and meteorological data through the Internet of Things (IoT) platform or cloud computing platform to form a unified data center. Use data processing software to automatically clean, filter and analyze the real-time collected data to form a visual hydrological and meteorological database to provide support for subsequent reservoir management and scheduling.

[0067] In the 3D virtual simulation construction module, long-term historical rainfall data is obtained from the meteorological stations in the area, covering daily rainfall, hourly rainfall intensity, etc. Large-scale rainfall data is obtained from meteorological satellites to supplement areas where the coverage of meteorological stations is insufficient. Commonly used satellites include NOAA, GPM, etc. The historical records of floods in the basin and the data on the corresponding rainfall conditions are collected to understand the characteristics of extreme rainfall events.

[0068] Arrange and clean rainfall data, remove abnormal data points, and ensure data accuracy and continuity. Interpolate data in time and space, supplement missing data, and generate continuous rainfall data sets. Convert processed rainfall data into a standard format that can be used for simulation (such as CSV, NetCDF).

[0069] Use global or regional climate models (e.g. GCM, WRF) to obtain future climate prediction data, with a particular focus on changes in rainfall patterns over the next few decades. Combined with climate change scenarios (e.g. RCP 4.5, RCP 8.5), predict changes in rainfall patterns under different greenhouse gas emission conditions.

[0070] Combining future weather forecasts with historical rainfall data to generate possible rainfall scenarios, the forecast data is refined by time period and spatial range to ensure that it matches the spatial resolution of existing hydrodynamic models.

[0071] Generate a 3D terrain model of the area around the reservoir using a digital elevation model (DEM) obtained by LiDAR or drones, ensuring that the model includes the dam, river, storage area, and drainage channels. Collect hydrological data such as historical flow, river sections, reservoir water levels, and flood discharge to set boundary conditions in the model.

[0072] Import terrain data obtained from LiDAR, drone aerial surveys, or satellite images into 3D modeling software (such as AutoCAD Civil 3D, Bentley ContextCapture, ArcGIS). Digitize the terrain features of the reservoir and its surrounding areas to ensure that the model reflects accurate surface elevation and morphology.

[0073] Use 3D modeling software to reconstruct the terrain of the reservoir area, including the reservoir dam, water storage area, flood discharge channel, etc. Ensure that the model is accurate enough to reflect the local details of the terrain, such as the width and slope of the river channel and the slope of the mountains around the reservoir.

[0074] Refine and optimize the 3D terrain model, remove unnecessary polygons or data redundancy, and ensure that the model can be used for simulation. Convert the 3D terrain model into a format acceptable to the hydrodynamic model, such as DEM or TIN file.

[0075] According to the project requirements, select the appropriate hydrodynamic modeling tool, such as HEC-RAS, MIKE 21, Delft3D, Flow3D, etc. Select a 1D, 2D or 3D hydrodynamic model, and determine the dimensionality requirements based on the size of the basin and the complexity of the reservoir (for example, large reservoirs usually require 2D or 3D simulations).

[0076] Import the generated 3D terrain model into the hydrodynamic modeling software. Set boundary conditions, including upstream inflow, downstream outlet location, dam section, spillway, etc. Import the hydrological data of the basin, including river flow, historical flood level and water storage level, etc., as the initial conditions of the model.

[0077] In the simulation model, the reservoir is divided into different functional areas, including water storage area, dam area, flood discharge area, drainage outlet and surrounding protection area, etc. According to the characteristics of each functional area, corresponding parameters are set, such as water depth, water flow velocity, flood discharge flow, etc.

[0078] Define key parameters of the model, such as Manning roughness (used to describe the flow resistance of the river channel), water viscosity, gravity acceleration, etc. Set the stress conditions of the dam body according to different water storage levels, taking into account water pressure and soil pressure.

[0079] Set different rainfall scenarios: Scenario 1: Conventional rainfall scenario: According to the historical average rainfall, set the conventional rainfall scenario to simulate the water storage and drainage behavior of the reservoir under normal rainfall conditions. Input the average rainfall intensity, inflow and evaporation data, and run the simulation in combination with the terrain model. Simulate the water level changes of the reservoir under the conventional water storage level and the normal operation of the drainage facilities.

[0080] Scenario 2: Extreme rainfall scenario: simulate extreme rainfall events that occur once in 100 years or once in 500 years to analyze the impact of floods on reservoirs. Set the extreme rainfall intensity and set the inflow under extreme rainfall scenarios based on historical flood data or meteorological forecast data. Simulate the water level changes after the peak flow enters the reservoir, the opening and drainage capacity of the spillway, and evaluate the peak flood carrying capacity of the reservoir.

[0081] Scenario 3: Future climate change scenarios: Based on the future climate change model, different climate scenarios are set, such as RCP4.5 or RCP 8.5, to simulate the impact of future climate change on reservoir water storage and drainage behavior. Import future rainfall forecast data, set the frequency and intensity of future extreme rainfall, and evaluate the reservoir's storage capacity under future climate conditions. Determine the reservoir's drainage capacity, flood management strategy, and possible changes in sediment deposition under future climate change scenarios.

[0082] Start the simulation with the hydrodynamic software and run each rainfall scenario step by step. Simulate the reservoir's water storage and drainage behavior, and record key parameters such as peak flow and flood discharge time. Export the simulation results to generate time series data of water level, flow, and pressure changes. Use visualization tools to generate a 3D dynamic display of water level changes, flood paths, storage areas, and drainage areas.

[0083] Sedimentation monitoring module: In the three-dimensional virtual simulation process, the sedimentation rate in the reservoir is monitored under different rainfall and flow conditions, the abnormal change trend of the sedimentation rate is determined, and the interference degree of sedimentation on the stability of the dam is evaluated.

[0084] Obtain data on sediment transport and deposition from historical observations, especially deposition after different flood events. Use river monitoring stations, flow meters or sediment sampling instruments to obtain long-term data on sediment transport in major rivers in the basin.

[0085] Use LiDAR or drone data to build a 3D terrain model of the reservoir bottom and surroundings to simulate the specific location of sediment deposition in the simulation. Obtain the flow velocity distribution at different water levels in the reservoir, especially the high flow velocity during flood season, to understand the hydrodynamic characteristics in the reservoir.

[0086] Different rainfall scenarios are set by combining historical rainfall, rainfall intensity and future climate forecast data. Hydrodynamic models (such as HEC-RAS, MIKE 21, Delft3D) are used to simulate the water flow rate of the reservoir under different flow conditions as the basic conditions for sediment deposition simulation.

[0087] Select an appropriate sediment transport and deposition model, commonly used models include Delft3D, MIKE 21, Flow3D, etc. These models can simulate the transport, suspension and deposition of sediment in water bodies. The model needs to support two-dimensional or three-dimensional simulation to capture the dynamic changes of sedimentation at the bottom of the reservoir.

[0088] Model input parameters: sediment particle characteristics (such as particle size, density, settling velocity). Reservoir flow velocity distribution, water depth changes, boundary conditions (including dam boundary and spillway flow). Input rainfall scenarios to generate sediment transport models under different flow and velocity conditions.

[0089] According to the geometric structure and functional zoning of the reservoir (such as water storage area, flood discharge area, and dam front area), the reservoir is divided into multiple simulation areas, and models are set up separately for the hydrodynamic characteristics of different areas.

[0090] Special attention should be paid to the main sediment transport channels of the reservoir (such as the water inlet and the main river mouth), which are high-risk areas for sediment deposition.

[0091] Boundary condition settings: Upstream boundary: Set the sediment input, particle size distribution and water flow rate. Downstream boundary: Set the drainage or flood discharge of the reservoir to simulate the process of sediment flowing out of the reservoir with water. Initial conditions: Set the initial state of the reservoir bottom (such as the initial sediment deposition thickness) and the reservoir water level.

[0092] Set the water flow and velocity under normal rainfall conditions based on historical average rainfall. Input sediment transport data based on flow and set the velocity range and sediment transport rate. Simulate the sediment deposition rate in the reservoir under normal water level and monitor the sediment thickness and regional distribution. Set the average rainfall and flow rate conditions for the next few decades based on historical data and climate forecasts. Simulate the sediment deposition behavior during long-term operation and predict the sediment deposition thickness in different areas of the reservoir. Assess the long-term impact of sediment deposition on the reduction of reservoir volume.

[0093] After the model is run, the time series data of sediment deposition rate are exported and the sediment thickness change graph is output, with special attention paid to the changes in sedimentation rate in the dam front area, near the spillway and at the bottom of the reservoir.

[0094] Through simulation data, the sediment deposition rate per unit time is calculated, and the sediment deposition rate of each area is compared with the data under different flow rates and water depths of the reservoir to generate the sediment deposition rate anomaly index, which is used to judge the abnormal change trend of the sediment deposition rate and evaluate the interference degree of sediment deposition on the stability of the dam body. The method for obtaining the sediment deposition rate anomaly index is as follows:

[0095] Through simulation monitoring, sediment deposition rate data (unit: mm / year or cm / year) is obtained in different areas of the reservoir. Sediment deposition rate data is time series data, which indicates the change of sediment deposition rate in different time periods (such as daily, weekly, and monthly). A representative period of time is selected as the initial data set. It is usually necessary to collect at least 20-25 data points to construct the initial control chart and calculate the average sediment deposition rate of the initial data set. , expressed as: ;in, is the sedimentation rate in the ith period, n is the total sample size, and the standard deviation of the sedimentation rate is calculated. , the formula is: ; Determine the control limits of the control chart. The control chart usually has three control lines, namely the center line CL, the upper control limit UCL and the lower control limit LCL, which can help determine whether the sediment deposition rate is abnormal. The center line is the average value of the sedimentation rate, and the upper control limit is set to the average value plus 3 times the standard deviation; the upper control limit indicates that the sedimentation rate is higher than the critical value of the normal fluctuation range, and points exceeding this value are considered to be abnormally high sedimentation rates. The lower control limit is set to the average value minus 3 times the standard deviation. The lower control limit indicates that the sedimentation rate is lower than the critical value of the normal fluctuation range, and points below this value are considered to be abnormally low sedimentation rates. If the calculated lower control limit LCL is a negative value, it can be set to zero because the sediment deposition rate cannot be negative. Draw a control chart, X-axis: represents the time series (such as daily, weekly, monthly and other time periods). Y-axis: represents the sediment deposition rate in each time period. Control lines: draw the center line CL, upper control limit UCL and lower control limit LCL, and plot the sediment deposition rate at each time point on the graph. When the sediment deposition rate data point in a certain period of time When it is greater than the upper control limit UCL or less than the lower control limit LCL, it is considered an abnormal point. If the sediment deposition rate data point is greater than the upper control limit UCL, it means that the sediment deposition rate is too fast and abnormally high deposition occurs. If the sediment deposition rate is less than the lower control limit LCL, it means that the sediment deposition rate is too low and there may be a problem. The sediment deposition rate anomaly index is marked as HK, and the sediment deposition rate data point When it is greater than the upper control limit UCL, ; When the sediment deposition rate data point When it is less than the lower control limit LCL, .

[0096] The larger the sediment deposition rate anomaly index is, the more the sedimentation rate significantly exceeds the normal fluctuation range, which means that the sediment accumulates rapidly in a short period of time. Rapid sedimentation may lead to a reduction in the reservoir's water storage capacity, blockage of the spillway, and even affect the stability of the dam. Excessively high sedimentation rates will increase the pressure in front of the dam, affect the stress distribution of the dam structure, and may cause local uneven settlement or dam instability. Therefore, the larger the anomaly index is, the greater the degree of interference of sediment deposition on the stability of the dam, the lower the accuracy of the three-dimensional simulation prediction results, and timely dredging or reinforcement measures are required.

[0097] On the contrary, the smaller the abnormal index is, the lower the sedimentation rate is, and there may even be sediment loss. Insufficient sedimentation or excessive sediment loss may lead to erosion of the riverbed and dam foundation, weaken the stability of the dam foundation, and thus cause safety hazards. Especially in the downstream area of ​​the dam, sediment loss will weaken its anti-scouring ability and increase the risk of dam damage. At this time, the insufficient sedimentation area should be monitored, and protective measures should be taken to prevent the long-term impact of excessive sediment loss on the safety of the dam.

[0098] Groundwater seepage simulation module: The groundwater seepage simulation tool is used to simulate the groundwater flow around the dam body, analyze the fluctuation amplitude of groundwater seepage pressure under different water levels and rainfall conditions, predict the risk of dam foundation sliding under different groundwater levels, and evaluate the influence of groundwater seepage on the stress distribution of the dam body.

[0099] Collect geological data around the dam, including soil type, rock structure, groundwater level changes, and permeability coefficient. These data can be obtained through geological exploration, drilling sampling, laboratory penetration testing, etc. Collect historical rainfall data, rainfall intensity, water level changes, and other information, especially hydrological data under extreme rainfall conditions. Long-term records of groundwater level changes in the dam area are also required. Obtain the geometric parameters of the dam, material properties (such as earth-rock dams, concrete dams), and structural design information of the dam foundation and dam abutments.

[0100] Use professional groundwater seepage simulation software, such as MODFLOW, SEEP / W or FEFLOW. These tools can simulate the flow and pressure distribution of groundwater under complex geological conditions. Create a three-dimensional model of the dam body and surrounding areas in the simulation software, covering key areas such as the dam body, dam foundation, downstream, and upstream. According to geological data, assign material properties (such as permeability coefficient, porosity, etc.) to different areas, and input dam material properties. Set the initial groundwater level, boundary conditions (such as upstream reservoir water level and downstream drainage conditions) and rainfall infiltration conditions. Divide the model into several finite elements (grids) to improve the accuracy of the calculation, especially using finer grid divisions in the dam foundation and high-risk areas of seepage.

[0101] Set the water level of the reservoir under normal water storage level to simulate the flow of groundwater around the dam body and dam foundation. The simulation model calculates the distribution of seepage pressure of the reservoir water level on the dam foundation, especially the pressure difference between the inside and outside of the dam body. Record the size and distribution of seepage pressure in the dam body and dam foundation area under normal water level conditions. Set the reservoir at extreme water level conditions at high water level (such as flood period) to simulate the impact of high water level on groundwater seepage pressure. Analyze the pressure rise in the dam foundation area and the change of seepage path when the dam body is at high water level, and identify whether there is an increase or concentration of seepage in the dam body. Compare the amplitude of groundwater seepage pressure fluctuation under normal water level and extreme water level conditions to find high-risk areas.

[0102] Simulate the change of groundwater level under normal rainfall conditions and evaluate the impact of soil water absorption around the dam on groundwater flow. Calculate the process of rainfall infiltration into groundwater based on rainfall and permeability coefficient, simulate the gradual rise of groundwater level and pressure change. Analyze the distribution of seepage pressure in the dam foundation and dam sidewall area during rainfall infiltration.

[0103] Simulate the rapid rise of groundwater level and pressure changes under extreme rainfall conditions (such as heavy rain and continuous rainfall for multiple days). Simulate the increase of groundwater flow velocity and changes in infiltration paths caused by extreme rainfall to identify possible leakage and piping in the dam body. Record the areas where the infiltration pressure rises rapidly under extreme rainfall conditions, especially in the high-risk areas at the junction of the dam foundation and the dam abutment.

[0104] Set different combination scenarios, such as high water level superimposed on heavy rain, to simulate the changes in seepage pressure of the dam under complex conditions. Analyze how the groundwater seepage pressure fluctuates under these scenarios and whether it will threaten the stability of the dam. Identify which areas have the largest pressure fluctuations under this combination scenario and determine the areas that need to be monitored and reinforced.

[0105] The seepage pressure data under normal water level and extreme water level, normal rainfall and extreme rainfall conditions are obtained, and the seepage pressure fluctuation amplitude of groundwater under different water level and rainfall conditions is analyzed to generate the seepage pressure fluctuation index, which is used to predict the risk of dam foundation sliding under different groundwater level conditions and evaluate the influence of groundwater seepage on the force distribution of the dam body. The method for obtaining the seepage pressure fluctuation index is as follows:

[0106] Obtain osmotic pressure data within a time period of W , including pressure values ​​under different water levels and rainfall conditions, and obtaining the prior distribution of seepage pressure based on historical data, expressed as the mean of the pressure and variance Based on the groundwater seepage model, the distribution form of the seepage pressure is defined, and the seepage pressure obeys the normal distribution ; Prior distribution represents the initial understanding of osmotic pressure when there is no new observation data, constructing the likelihood function, the likelihood function Describes the possibility of observing the current osmotic pressure data under given parameter conditions. The expression is: ; is the ith observed osmotic pressure data point, and m is the number of observed data. Apply Bayes’ theorem to calculate the posterior distribution. Updated by prior distribution and likelihood function, the expression is: ; is the prior distribution of groundwater pressure, is the marginal likelihood of the observed data; update the parameters and obtain the posterior distribution through Bayes’ theorem, the expressions are: ; ; In the formula, is the updated mean osmotic pressure, is the updated osmotic pressure variance, is the variance of the observed data, and the osmotic pressure fluctuation index is calculated as follows: ; Where GH is the osmotic pressure fluctuation index.

[0107] The larger the seepage pressure fluctuation index, the greater the variation of groundwater seepage pressure under different water levels and rainfall conditions, indicating that groundwater infiltration has generated significant dynamic stress in the dam foundation and dam body area. Large fluctuations in seepage pressure may lead to a weakening of the shear strength of the dam body material, especially in areas where the seepage path is concentrated or pressure accumulates, which may easily lead to local dam foundation sliding or leakage risks. Such fluctuations will change the force distribution of the dam body and increase local stress concentration, thereby threatening the overall stability of the dam body and causing a high risk of sliding.

[0108] On the contrary, a small seepage pressure fluctuation index means that the groundwater pressure changes less under different water levels and rainfall conditions, and the seepage pressure is relatively stable. This situation usually indicates that the impact of groundwater infiltration on the dam is limited, the force distribution is relatively uniform, and the risk of sliding is low. Although small pressure fluctuations may mean that the dam is in a safe state, it is still necessary to monitor long-term change trends, especially during extreme rainfall or rapid water level rise, to ensure that the dam is always in a stable operating state.

[0109] Simulation accuracy evaluation module: Comprehensively analyze the interference degree of sediment deposition on the stability of the dam body and the influence of groundwater infiltration on the stress distribution of the dam body, and determine the accuracy of virtual simulation of the stress conditions of the reservoir dam body under different operating conditions based on the analysis results.

[0110] The sediment deposition rate anomaly index and the seepage pressure fluctuation index are converted into the first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses each group of first eigenvectors to predict the simulation accuracy value label of the virtual simulation of the stress conditions of the reservoir dam under different operating conditions as the prediction target, and takes minimizing the sum of the prediction errors of the simulation accuracy value labels of all virtual simulations of the stress conditions of the reservoir dam under different operating conditions as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The simulation accuracy value of the virtual simulation of the stress conditions of the reservoir dam under different operating conditions is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0111] The method for obtaining the simulation accuracy value of the stress condition of the reservoir dam under different operating conditions by virtual simulation is as follows: from the first eigenvector training data of the trained machine learning model, the corresponding function expression is obtained: ; In the formula, is the output function of the model, HK is the sediment deposition rate anomaly index, GH is the seepage pressure fluctuation index, It is the simulation accuracy value of the virtual simulation of the stress conditions of the reservoir dam under different operating conditions.

[0112] Simulation result division module: divides the simulation accuracy of the virtual simulation on the stress conditions of the reservoir dam under different operating conditions into accurate simulation results, incompletely accurate simulation results and inaccurate simulation results, and manages them accordingly.

[0113] The obtained simulation accuracy value of the virtual simulation on the stress condition of the reservoir dam body under different operating conditions is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the simulation accuracy value of the virtual simulation on the stress condition of the reservoir dam body under different operating conditions is compared with the first standard threshold and the second standard threshold respectively;

[0114] If the simulation accuracy value of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is greater than the second standard threshold value, it means that the simulation accuracy of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is high, and a high-accuracy model signal is generated at this time, and the simulation results of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions are classified as accurate simulation results;

[0115] If the simulation accuracy value of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the simulation accuracy of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is general, and a medium accuracy model signal is generated at this time, and the simulation results of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions are classified as incompletely accurate simulation results;

[0116] If the simulation accuracy value of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is less than the first standard threshold, it means that the simulation accuracy of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is low. At this time, a low-accuracy model signal is generated, and the simulation results of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions are divided into inaccurate simulation results, and the divided simulation results with different accuracies are processed accordingly.

[0117] Management module: For accurate simulation results, the management module makes decisions and scheduling strategies directly based on the simulation results; for inaccurate simulation results, a warning should be triggered immediately and the simulation model should be recalibrated.

[0118] For accurate simulation results, the management module can make decisions and dispatch strategies directly based on the simulation results to ensure the safety of reservoir operation and dam body. According to the simulation results, confirm whether the current operating conditions meet the safety requirements. For example, under high water level or extreme rainfall conditions, whether the simulation shows that the structure of the reservoir dam body is stable. Based on the simulation results, the management module automatically generates dispatch strategies. For example: adjust the reservoir water storage or flood discharge strategy. Adjust the dam reinforcement or maintenance plan, such as whether the water level needs to be lowered to reduce the stress on the dam body. The optimized decision is passed to the reservoir automation control system to perform specific operations. For example, open the spillway, adjust the water storage level, or start the sediment cleaning equipment in the reservoir. After executing the decision, the management module continuously monitors the operation through real-time sensors (monitoring water level, seepage pressure, etc.) and verifies the accuracy of the simulation prediction based on the new data. If the subsequent monitoring results are consistent with the simulation prediction, the current dispatch strategy will continue to be executed; if there is a deviation, the system will re-simulate and adjust.

[0119] When the simulation results are inaccurate, the management module should trigger a warning and recalibrate the simulation model to avoid making wrong decisions based on inaccurate results. The system immediately generates a low-accuracy model signal and automatically triggers the early warning mechanism. At this time, the manager will receive a warning notification from the system, indicating that the current simulation results may not be credible and decisions cannot be made directly based on the simulation. At the same time, strengthen the real-time monitoring of the reservoir dam body, pay close attention to key indicators (such as dam body force, seepage pressure, and water storage) to prevent the expansion of safety risks. Check whether there are errors in the simulation input data, especially the sediment deposition rate and seepage pressure fluctuation data. By comparing historical data, check whether the sensor is working properly and whether the data is reliable.

[0120] The latest data on seepage pressure, water level, sediment deposition rate, etc. are obtained from the real-time monitoring system of the reservoir operation to calibrate the parameters of the simulation model. According to the new monitoring data, the parameter settings of the simulation model are readjusted, including: correction of permeability coefficient. Update of groundwater level and flow rate. Real-time update of sediment deposition rate. Based on the updated data and parameters, the simulation model is re-run to simulate the stress conditions of the current reservoir dam body. After re-simulation, the accuracy of the new simulation results is evaluated again using the machine learning model. If the new simulation results pass the evaluation and the accuracy value exceeds the threshold, the warning is lifted and normal scheduling is resumed. If there is still a large error after re-simulation, it is necessary to further check the input data source or consider a more complex simulation model, such as introducing a more accurate geological simulation or hydrodynamic model. In the case of unreliable simulation results, emergency strategies are formulated based on real-time monitoring data and experience. For example: lower the reservoir water level to reduce the pressure on the dam body. Suspend or delay high-risk reservoir scheduling operations and make decisions after the simulation results are calibrated. Through the feedback of real-time data, the simulation model is continuously iterated and updated until the accuracy of the simulation results meets the standard requirements.

[0121] Dynamic adjustment module: For inaccurate simulation results, the simulation accuracy of the virtual simulation of the stress conditions of the reservoir dam under different operating conditions within a fixed time period is deeply analyzed, and the abnormal degree of simulation deviation is judged according to the analysis results. If the abnormal degree is high, the water storage capacity and dam bearing capacity are dynamically adjusted.

[0122] For inaccurate simulation results, set the simulation prediction value To simulate the predicted value of the dam body stress condition within time t, the actual monitoring value The actual monitoring results show the stress of the dam body; the simulation deviation represents the difference between simulation prediction and actual monitoring and is defined as: ; Calculate the deviation rate, the expression is: ; Indicates the percentage deviation of the simulation result from the actual monitoring value; the deviation rate in the subsequent fixed time period T is accumulated and summed to calculate the cumulative deviation in the T time period , the expression is: ; The cumulative deviation is used to determine whether the simulation deviation persists for a long time, so as to evaluate the degree of abnormality. Based on the cumulative deviation and deviation rate, the abnormality index is calculated. To evaluate the severity of simulation deviation, the expression is: ; In the formula, It is the average value of the actual force value in the time period T. The higher the anomaly index, the larger the simulation error is over a longer period of time and the farther it deviates from the normal value.

[0123] If the abnormal index , it is necessary to dynamically adjust the reservoir water storage capacity, is the preset abnormal index reference threshold, and the current water storage capacity is set to , the maximum safe water storage capacity is , according to the degree of abnormality, the formula for dynamically adjusting the water storage capacity is: ;in, is the adjusted water storage capacity, α is the adjustment coefficient, usually between 0<α≤1, and is set according to the specific reservoir safety requirements. As the abnormal index increases, the water storage capacity will decrease accordingly.

[0124] Similarly, based on the abnormal index, the bearing capacity of the dam is adjusted . Assume the current carrying capacity is , the formula for adjusting the carrying capacity under abnormal conditions is: ;in, is the adjusted bearing capacity of the dam, and β is the bearing capacity adjustment factor, which depends on the material properties of the dam structure and its safety margin. As the anomaly index increases, the bearing capacity will decrease, requiring more stringent control and reinforcement measures.

[0125] For the safe operation range of the dam body, the upper limits of water storage and carrying capacity are redefined through simulation adjustment. If the anomaly persists, the upper limit of water storage will be further reduced. and carrying capacity limit , the expressions are: ; ; Among them, γ and δ are control coefficients for long-term adjustment, which are set according to specific safety standards. For maximum safe water storage capacity, is the maximum bearing capacity of the dam.

[0126] In this embodiment, the data monitoring module collects three-dimensional terrain and hydrometeorological data in different time periods by dividing the project area into several sub-areas; the three-dimensional virtual simulation construction module sets rainfall scenarios and simulates reservoir water storage and drainage behaviors according to historical rainfall and meteorological forecasts; the sediment deposition monitoring module monitors the sediment deposition rate and evaluates its impact on the stability of the dam body; the groundwater infiltration simulation module simulates groundwater flow and predicts the risk of dam foundation sliding under different water levels and rainfall conditions; the simulation accuracy evaluation module comprehensively analyzes the impact of sediment deposition and groundwater infiltration and evaluates the simulation accuracy; the simulation result classification module divides the simulation results into accurate, incompletely accurate and inaccurate, and takes different management measures respectively; the management module makes decisions based on accurate simulation results, issues warnings for inaccurate simulation results and recalibrates; the dynamic adjustment module dynamically adjusts the water storage capacity and dam bearing capacity in the incompletely accurate simulation results if the degree of abnormality is high.

[0127] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0128] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0129] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. Intelligent management system for water conservancy projects, characterized by: It includes data monitoring module, 3D virtual simulation construction module, sediment deposition monitoring module, groundwater infiltration simulation module, simulation accuracy evaluation module, simulation result division module, management module and dynamic adjustment module; Data monitoring module: divide the water conservancy project area into several sub-project areas, monitor and obtain the three-dimensional terrain data and hydrological and meteorological data of each sub-project area in different time periods; 3D virtual simulation construction module: Build a 3D virtual simulation model, set different rainfall scenarios based on historical rainfall data and future weather forecasts, and input different rainfall scenarios into the hydrodynamic model to simulate reservoir storage and drainage behavior; Sedimentation monitoring module: In the 3D virtual simulation process, the sedimentation rate in the reservoir is monitored under different rainfall and flow conditions, the abnormal change trend of sedimentation rate is determined, and the interference degree of sedimentation on the stability of the dam is evaluated; The sediment deposition monitoring module includes: generating a sediment deposition rate anomaly index after comparing the sediment deposition rate of each area with the data under different flow rates and different water depths of the reservoir. The method for obtaining the sediment deposition rate anomaly index is: The sediment deposition rate data of different areas of the reservoir are obtained through simulation monitoring. The sediment deposition rate data is time series data, which indicates the change of sediment deposition rate in different periods. The initial data set is selected and the average sediment deposition rate of the initial data set is calculated. , expressed as: ;in, is the sedimentation rate in the ith period, n is the total sample size, and the standard deviation of the sedimentation rate is calculated. , the formula is: ; Determine the control limits of the control chart. The control chart has three control lines, namely the center line CL, the upper control limit UCL and the lower control limit LCL. The center line is the average value of the sedimentation rate. The upper control limit is set to the average value plus 3 times the standard deviation; the lower control limit is set to the average value minus 3 times the standard deviation. Draw a control chart with the X-axis as the time series and the Y-axis as the sediment deposition rate for each period. The sediment deposition rate at each time point is plotted on the chart. When the sediment deposition rate data point When it is greater than the upper control limit UCL or less than the lower control limit LCL, it is considered an abnormal point; the sediment deposition rate anomaly index is marked as HK, and the sediment deposition rate data point When it is greater than the upper control limit UCL, ; When the sediment deposition rate data point When it is less than the lower control limit LCL, ; Groundwater seepage simulation module: Use the groundwater seepage simulation tool to simulate the groundwater flow around the dam body, analyze the fluctuation amplitude of groundwater seepage pressure under different water levels and rainfall conditions, predict the risk of dam foundation sliding under different groundwater levels, and evaluate the impact of groundwater seepage on the force distribution of the dam body; The groundwater infiltration simulation module includes: obtaining the infiltration pressure data under normal water level and extreme water level, normal rainfall and extreme rainfall conditions, and generating the infiltration pressure fluctuation index after analyzing the fluctuation amplitude of the infiltration pressure of groundwater under different water level and rainfall conditions. The method for obtaining the infiltration pressure fluctuation index is: Obtain osmotic pressure data within a time period of W , including pressure values ​​under different water levels and rainfall conditions, and obtaining the prior distribution of seepage pressure based on historical data, expressed as the mean of the pressure and variance Based on the groundwater seepage model, the distribution form of the seepage pressure is defined, and the seepage pressure obeys the normal distribution ; Prior distribution represents the initial understanding of osmotic pressure when there is no new observation data, constructing the likelihood function, the likelihood function Describes the possibility of observing the current osmotic pressure data under given parameter conditions. The expression is: ; is the ith observed osmotic pressure data point, and m is the number of observed data. Apply Bayes’ theorem to calculate the posterior distribution. Updated by prior distribution and likelihood function, the expression is: ; is the prior distribution of groundwater pressure, is the marginal likelihood of the observed data; update the parameters and obtain the posterior distribution through Bayes’ theorem, the expressions are: ; ; In the formula, is the updated mean osmotic pressure, is the updated osmotic pressure variance, is the variance of the observed data, and the osmotic pressure fluctuation index is calculated as follows: ; Where GH is the osmotic pressure fluctuation index; Simulation accuracy evaluation module: Comprehensively analyze the interference degree of sediment deposition on the stability of the dam body and the influence of groundwater infiltration on the stress distribution of the dam body, and determine the simulation accuracy of the virtual simulation of the stress conditions of the reservoir dam body under different operating conditions based on the analysis results; The simulation accuracy evaluation module includes: converting the sediment deposition rate anomaly index and the seepage pressure fluctuation index into a first eigenvector, taking the first eigenvector as the input of the machine learning model, the machine learning model uses each group of first eigenvectors to predict the simulation accuracy value label of the stress condition of the reservoir dam body under different operating conditions of the virtual simulation as the prediction target, and minimizing the sum of the prediction errors of the simulation accuracy value labels of the stress condition of the reservoir dam body under different operating conditions of all virtual simulations as the training target, training the machine learning model until the sum of the prediction errors reaches convergence and stops the model training, and determines the simulation accuracy value of the virtual simulation of the stress condition of the reservoir dam body under different operating conditions according to the model output results, wherein the machine learning model is a polynomial regression model; Simulation result division module: divides the simulation accuracy of the virtual simulation on the stress conditions of the reservoir dam under different operating conditions into accurate simulation results, incompletely accurate simulation results and inaccurate simulation results, and manages them accordingly; Management module: For accurate simulation results, the management module directly makes decisions and scheduling strategies based on the simulation results; for inaccurate simulation results, a warning should be triggered immediately and the simulation model should be recalibrated; Dynamic adjustment module: For inaccurate simulation results, the simulation accuracy of the virtual simulation of the stress conditions of the reservoir dam under different operating conditions within a fixed time period is deeply analyzed, and the abnormal degree of simulation deviation is judged according to the analysis results. If the abnormal degree is high, the water storage capacity and dam bearing capacity are dynamically adjusted.

2. The intelligent management system for water conservancy projects according to claim 1 is characterized in that: In the simulation result division module, the simulation accuracy of the virtual simulation on the stress conditions of the reservoir dam under different operating conditions is divided into accurate simulation results, incompletely accurate simulation results and inaccurate simulation results, specifically: The obtained simulation accuracy value of the virtual simulation on the stress condition of the reservoir dam body under different operating conditions is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the simulation accuracy value of the virtual simulation on the stress condition of the reservoir dam body under different operating conditions is compared with the first standard threshold and the second standard threshold respectively; If the simulation accuracy value of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is greater than the second standard threshold value, it means that the simulation accuracy of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is high, and a high-accuracy model signal is generated at this time, and the simulation results of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions are classified as accurate simulation results; If the simulation accuracy value of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the simulation accuracy of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is general, and a medium accuracy model signal is generated at this time, and the simulation results of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions are classified as incompletely accurate simulation results; If the simulation accuracy value of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is less than the first standard threshold, it means that the simulation accuracy of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions is low. At this time, a low-accuracy model signal is generated, and the simulation results of the virtual simulation on the stress conditions of the reservoir dam body under different operating conditions are classified as inaccurate simulation results.

3. The intelligent management system for water conservancy projects according to claim 1 is characterized by: In the dynamic adjustment module, for inaccurate simulation results, the simulation accuracy of the virtual simulation of the stress conditions of the reservoir dam under different operating conditions within a fixed time period is deeply analyzed, and the abnormal degree of simulation deviation is judged according to the analysis results, specifically: For inaccurate simulation results, set the simulation prediction value To simulate the predicted value of the dam body stress condition within time t, the actual monitoring value The actual monitoring results show the stress of the dam body; the simulation deviation represents the difference between simulation prediction and actual monitoring and is defined as: ; Calculate the deviation rate, the expression is: ; is the deviation rate, which indicates the percentage deviation of the simulation result from the actual monitoring value; The deviation rate in the subsequent fixed time period T is accumulated and summed to calculate the cumulative deviation in the T time period , the expression is: ; According to the cumulative deviation and deviation rate, calculate the abnormality index, the expression is: , the expression is: ; In the formula, is the average value of the actual force value in the time period T.

4. The intelligent management system for water conservancy projects according to claim 3 is characterized by: If the abnormal index , it is necessary to dynamically adjust the reservoir water storage capacity, is the preset abnormal index reference threshold, and the current water storage capacity is set to , the maximum safe water storage capacity is , according to the degree of abnormality, the formula for dynamically adjusting the water storage capacity is: ; where α is the adjustment coefficient, is the adjusted water storage capacity; Adjust the dam bearing capacity based on the anomaly index , assuming the current carrying capacity is , the formula for adjusting the carrying capacity under abnormal conditions is: ; where β is the load-bearing capacity adjustment factor, is the adjusted bearing capacity of the dam; For the safe operation range of the dam body, the upper limits of water storage and carrying capacity are redefined through simulation adjustment. If the anomaly persists, the upper limit of water storage will be further reduced. and carrying capacity limit , the expressions are: ; ; where γ and δ are control coefficients for long-term adjustment, For maximum safe water storage capacity, is the maximum bearing capacity of the dam.

Images