Method and system for adjusting water level of hydro-junction

Abstract

The invention provides a hydro-junction water level adjusting method and system. Wherein the sediment deposition space-time atlas containing the sedimentary gradient is constructed and periodically corrected through scanning of an underwater robot and synchronous acquisition of satellite remote sensing data. A radar array is deployed at a bend section of a flood discharge channel, and vector distribution of a three-dimensional sediment flow velocity field is analyzed by using multi-frequency pulse signals. Through correlation analysis of the dynamic coupling sedimentary gradient and the flow velocity field, a correlation area of flow velocity sudden change and sediment deposition is accurately identified, and a constraint condition set including coordinate positioning and a flow-sediment matching rule is established. And in combination with real-time water level monitoring data, calculating and generating a dynamic adjustment parameter sequence of the opening degree of the flood discharge gate by using a constraint condition set, thereby realizing accurate regulation and control of the water level. According to the technical scheme, the flood discharge parameters are intelligently optimized, the water level of the hydro-junction can be accurately regulated and controlled, the sediment deposition problem of a high-sediment-content water area is effectively relieved, and safe operation and flood control efficiency of the junction are guaranteed.

Description

Technical Field

[0001] The present application relates to the technical field of automated control of water conservancy projects, and in particular to a method and system for regulating the water level of a water conservancy hub. Background Art

[0002] In the operation of hydropower hubs on rivers with high sediment concentrations, the ability of water to carry sediment is closely linked to the dynamic changes in water levels. This requires precise regulation to achieve the coordinated goals of controlling sediment accumulation in the reservoir area and stabilizing the water level. These scenarios require hydropower hubs to monitor sediment transport status in real time and dynamically adjust discharge strategies to avoid problems such as siltation in front of the gates, loss of reservoir capacity, and severe scouring of the downstream riverbed, while ensuring flood control safety, power generation efficiency, and ecological needs.

[0003] Currently, the main technical solution to this problem is an automated control system that combines water-sediment dynamics models with real-time monitoring data. This system continuously collects water level, flow velocity, and sediment concentration data through underwater sensors. It then uses numerical models coupled with water-sediment transport mechanisms to predict short-term sedimentation trends. Based on these predictions, it generates optimized instructions for gate opening and closing or unit operation.

[0004] However, existing solutions still face several bottlenecks in practical application. The accuracy of the sand coupling model is limited by the ability to obtain real-time parameters such as sediment particle size distribution. This is particularly true under extreme sediment concentrations, which can lead to model deviations, resulting in a mismatch between control instructions and actual operating conditions. Furthermore, sensors are prone to signal attenuation or data drift in highly turbid waters, and the absence of key parameters can cause control logic failures. Summary of the Invention

[0005] The present application provides a method and system for regulating the water level of a water conservancy hub, which is used to solve the problem of limiting the robustness and universality of the system in high sediment content scenarios in the existing technology.

[0006] In a first aspect, the present application provides a method for regulating the water level of a water conservancy hub, comprising: For high-sediment-content waters in the reservoir area of ​​the water conservancy hub and the downstream section of the spillway, multi-dimensional scanning data of the riverbed topography by underwater robots and global coverage data from satellite remote sensing are simultaneously obtained to generate a spatiotemporal map of sediment deposition that includes a sediment gradient. This spatiotemporal map is then periodically revised based on changes in the sediment gradient. A radar array is deployed in the bend transition section of the spillway outlet. Based on the modified spatiotemporal map of sediment deposition, the corresponding three-dimensional sediment velocity field vector distribution is calculated in real time by using the scattering intensity differences and phase shift trajectories of multi-band pulse signals. Dynamically coupling the sediment gradient with the three-dimensional sediment velocity field vector distribution to identify the associated region of velocity mutation and sediment deposition within the bend transition section, and generating a constraint condition set including the coordinates of the associated region and the flow velocity and sediment matching rule; According to the set of constraints, combined with the real-time water level monitoring data and the target water level range of the water conservancy hub reservoir area, a dynamic adjustment parameter sequence of the flood discharge gate opening is calculated.

[0007] Optionally, the time series data of the sediment gradient and the three-dimensional sediment velocity field vector distribution are spatially superimposed based on the geographic coordinate system of the bend transition section to generate a synchronously updated sediment velocity coupled data set; In the sediment-flow velocity coupled data set, detecting a velocity vector modulus mutation region in the three-dimensional sediment velocity field that meets the sedimentation matching rule, and marking the velocity vector modulus mutation region as a primary velocity mutation region; superimposing the second-order derivative distribution of the sedimentation gradient on the primary velocity mutation area to screen out areas that simultaneously meet the extreme value judgment rule and where the sedimentation gradient undergoes a mutation as velocity-deposition associated areas; The three-dimensional geographic coordinate set of the velocity-deposition associated area is extracted, and the numerical correspondence between the sediment gradient and the velocity vector modulus in the velocity-deposition associated area is counted. According to the numerical correspondence, a constraint condition set including the coordinates of the associated area and the velocity and sediment matching rules is generated.

[0008] Optionally, based on the velocity vector length distribution of the spatial unit in the primary velocity mutation zone, a historical benchmark and a current cycle benchmark of the velocity vector length are set, and a primary screening area is delineated in the spatial unit that satisfies both the historical benchmark and the current cycle benchmark; Extracting sediment gradient data for each location point in the primary screening area, and superimposing and calculating the second-order derivative distribution of the sediment gradient in adjacent time periods for each location point; Setting parallel virtual measurement lines in the primary screening area along the water flow direction, generating a discrete point sequence at array intervals for each virtual measurement line, and mapping the second-order derivative distribution calculation results corresponding to the discrete point sequence into a second-order derivative intensity distribution; Based on the overall distribution state of the second-order derivative intensity distribution, an extreme value determination rule is set, and adjacent discrete point areas that continuously meet the extreme value determination rule are marked as sedimentation gradient mutation areas; The sediment gradient mutation zone is superimposed in three-dimensional space along the vertical direction of the virtual measurement line, and the spatial coordinates of the primary screening area and the distribution range of the sediment gradient mutation zone are integrated. The set of superimposed spatial units that simultaneously meet the velocity vector length benchmark condition and the extreme value judgment rule is defined as the velocity deposition association area.

[0009] Optionally, a multi-band pulse signal including a low-frequency penetration signal and a high-frequency tracking signal is synchronously transmitted, and after receiving the reflected signal, a separation is performed to generate a density distribution map of the sediment penetrating to the bottom of the riverbed and a movement trajectory map of the surface sediment; Extracting the sediment accumulation density change rate of each depth layer in the vertical direction from the sediment density distribution map, and combining it with the sediment amount gradient at the corresponding position in the revised sediment deposition time-space map to generate a vertical density gradient correlation profile; Performing three-dimensional spatial decomposition on the surface sediment movement trajectory map, and establishing a surface particle movement direction angle distribution map based on the spatial position offset of the reflection signal feature points within the continuous transmission cycle; The vertical density gradient correlation profile is spatially matched with the surface particle movement direction angle distribution map, and the surface movement direction angle data of the corresponding position is superimposed in the vertical density gradient mutation area. The corresponding three-dimensional sediment velocity field vector distribution is generated through vector direction transfer.

[0010] Optionally, spatially matching each depth layer data of the vertical density gradient correlation profile with the plane coordinates of the surface particle movement direction angle distribution map is performed to screen out the vertical density gradient mutation area corresponding to the plane coordinates; Extracting surface motion direction angle data of the plane coordinates corresponding to the vertical density gradient mutation area, and calculating the horizontal motion vector of each plane coordinate point based on the surface motion direction angle data; Determine the gradient direction change amount of each position point in the vertical density gradient mutation area, and calculate the corresponding vertical direction correction value based on the gradient direction change amount; After superimposing the vertical direction correction value on the horizontal motion vector, the vector is transferred layer by layer along the vertical density gradient change direction, and the vector direction is corrected based on the offset angle between the gradient direction and the surface direction angle; The corrected motion vectors of all depth layers in the vertical density gradient mutation area are spatially superimposed to generate a three-dimensional sediment velocity field vector distribution covering the corresponding plane coordinates.

[0011] Optionally, the underwater robot is set to scan the riverbed topography, and the sediment accumulation thickness and multi-dimensional scanning data of each scanning point are simultaneously obtained, and the full-area coverage data of satellite remote sensing is simultaneously obtained during the robot scanning interval; The thickness variation difference of the sediment accumulation thickness at each scanning point is calculated according to the continuous scanning cycle, and the ratio of the thickness variation difference to the horizontal spacing between adjacent scanning points is used as the sediment gradient. The multi-dimensional scanning data and the global coverage data are combined to generate a spatiotemporal map of sediment deposition including the gradient direction and sediment gradient; When the direction of the deposition gradient at any scanning point continuously deviates from the initial direction by more than a set angle, or the fluctuation amplitude of the deposition gradient value exceeds a set ratio, the scanning path encryption of the area where the point is located is triggered; The sediment accumulation thickness is reacquired in the encrypted scanning area, and the value and direction of the sediment gradient are corrected based on the new sediment accumulation thickness data, overwriting the data of the corresponding area in the original space-time map.

[0012] Optionally, real-time water level monitoring data of the water conservancy hub reservoir area is obtained, the deviation between the current water level and the median of the target water level interval is calculated, and the intensity of the water level change trend in the future period is predicted based on the historical water level change data; Analyzing the flow velocity threshold and the sedimentation gradient threshold of each associated area in the constraint condition set, converting the ratio of the current measured flow velocity to the flow velocity threshold into a flow velocity adjustment parameter, and converting the ratio of the current sedimentation gradient to the sedimentation gradient threshold into a sedimentation gradient adjustment parameter; Performing a weighted fusion of the flow velocity adjustment parameter and the sedimentation gradient adjustment parameter to generate initial adjustment parameter values ​​for each associated area, and applying a directional correction to the initial adjustment parameter values ​​based on the water level deviation direction and trend strength; The corrected adjustment parameter values ​​of each area are arranged in time series to ensure that the parameter changes in adjacent time periods do not exceed the safety threshold of the gate mechanical action. The parameter differences between spatially adjacent areas are smoothed to generate a dynamic adjustment parameter sequence for the flood discharge gate opening.

[0013] In a second aspect, the present application provides a water level regulation system for a water conservancy hub, comprising: An acquisition module, which simultaneously acquires multi-dimensional scanning data of the riverbed topography from underwater robots and global coverage data from satellite remote sensing for high-sediment-content waters in the reservoir area of ​​the water conservancy hub and the downstream river section of the spillway, to generate a spatiotemporal map of sediment deposition including a sediment gradient, and periodically modifies the spatiotemporal map of sediment deposition based on changes in the sediment gradient; A calculation module deploys a radar array in the bend transition section of the spillway outlet. Based on the modified spatiotemporal map of sediment deposition, the module calculates the corresponding three-dimensional sediment velocity field vector distribution in real time by using the scattering intensity differences and phase shift trajectories of multi-band pulse signals. a generation module that dynamically couples the sediment gradient with the three-dimensional sediment velocity field vector distribution, identifies the associated region of velocity mutation and sediment deposition within the bend transition section, and generates a constraint condition set including the coordinates of the associated region and a matching rule between velocity and sediment; The calculation module calculates the dynamic adjustment parameter sequence of the flood discharge gate opening according to the set of constraints and in combination with the real-time water level monitoring data and the target water level range of the water conservancy hub reservoir area.

[0014] In a third aspect, an embodiment of the present application provides a computing device comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for regulating the water level of a water conservancy hub as described in the first aspect above.

[0015] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program. When the computer program is executed by a computer, the method for regulating the water level of a water conservancy hub as described in the first aspect is implemented.

[0016] In an embodiment of the present application, for high-sediment-content waters in the reservoir area of ​​a water conservancy hub and the downstream section of a spillway, multi-dimensional scanning data of the riverbed topography by an underwater robot and global coverage data from satellite remote sensing are simultaneously acquired to generate a spatiotemporal map of sediment deposition including a sediment gradient, and the spatiotemporal map of sediment deposition is periodically revised based on changes in the sediment gradient. A radar array is deployed in the bend transition section of the spillway outlet. Based on the revised spatiotemporal map of sediment deposition, the corresponding three-dimensional sediment velocity field vector distribution is calculated in real time by using the scattering intensity differences and phase offset trajectories of multi-band pulse signals. Dynamic coupling analysis is performed between the sediment gradient and the three-dimensional sediment velocity field vector distribution to identify regions within the bend transition section where velocity mutations are associated with sediment deposition, and a constraint condition set is generated including the coordinates of the associated regions and flow velocity and sediment matching rules. Based on the constraint condition set, a dynamic adjustment parameter sequence for the spillway gate opening is calculated in combination with real-time water level monitoring data of the water conservancy hub reservoir and a target water level range.

[0017] This application has the following beneficial effects: This application constructs a high-precision spatiotemporal map of sediment deposition through multi-source data fusion (underwater robot scanning and satellite remote sensing), combines the multi-band signals of the radar array to analyze the three-dimensional sediment velocity field, dynamically couples the sediment gradient and velocity field vector, accurately identifies the associated areas of velocity mutation in the bend section and sediment deposition, and intelligently calculates the sequence of flood discharge gate control parameters based on a set of constraint rules and real-time water level data, ultimately achieving precise prevention and control of sediment deposition and dynamic optimization of water levels in waters with high sediment content, significantly improving the flood control safety, reservoir capacity stability and sediment discharge efficiency of the water conservancy hub, while taking into account automated decision-making and multi-objective collaborative control capabilities.

[0018] Furthermore, by spatially overlaying sediment gradient time series data with the three-dimensional sediment velocity field vector based on the geographic coordinate system of the bend transition section, a dynamically updated sediment-velocity coupled dataset was generated. Based on the sediment matching rule, regions with sudden changes in velocity vector modulus length in the dataset were detected and marked as primary velocity mutation regions. Furthermore, the second-order derivative distribution of the sediment gradient was overlaid to screen out overlapping regions of sediment gradient mutation that simultaneously met the extreme value conditions. Finally, their three-dimensional coordinates and sediment-velocity numerical relationship were extracted to generate a set of constraints containing matching rules. This significantly improved the spatiotemporal resolution and decision-making reliability of sedimentation risk positioning, thereby optimizing the targeted sediment discharge and regulation capabilities of the flood discharge gates, ensuring the accuracy of water level control and the safety of hub operations.

[0019] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0021] Figure 1 A flow chart showing a method for regulating the water level of a water conservancy hub provided by the present application is shown; Figure 2 A schematic structural diagram of a water level regulation system for a water conservancy project provided by the present application is shown; Figure 3 A schematic structural diagram of a computing device provided by the present application is shown. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0023] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to being different types.

[0024] Researchers have found that due to the coupling of dynamic sediment deposition and complex flow fields, traditional water level control methods for water conservancy hubs in high-sediment-content waters suffer from problems such as insufficient sediment monitoring accuracy, a lack of velocity-sedimentation correlation modeling, and delayed gate control. This leads to reservoir capacity loss, low sediment discharge efficiency, and increased flood control risks. Based on this, a multi-source data-driven intelligent water level regulation method for water conservancy hubs is proposed. This method constructs a dynamic sediment deposition map through the fusion of underwater robot scanning and satellite remote sensing, combines a radar array to calculate the three-dimensional velocity field in real time, and generates targeted constraint rules based on the dynamic coupling analysis of sedimentation and velocity, ultimately achieving precise temporal control of gate openings.

[0025] The technical solution of this application can be applied to the management of water conservancy hubs in rivers with high sediment content, and is especially suitable for sediment prevention and control and water level optimization control in complex flow scenarios such as spillway bend transition sections and reservoir siltation-sensitive areas.

[0026] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0027] Figure 1 A flow chart of a method for regulating the water level of a water conservancy hub is provided for the embodiment of the present application, as shown in FIG. Figure 1 As shown, the method includes: 101. For high-sediment-content waters in the reservoir area of ​​the water conservancy hub and the downstream river section of the spillway, simultaneously obtain multi-dimensional scanning data of the riverbed topography from underwater robots and full-area coverage data from satellite remote sensing to generate a spatiotemporal map of sediment deposition that includes a sediment gradient, and periodically revise the spatiotemporal map of sediment deposition based on changes in the sediment gradient; In step 101, the reservoir area of ​​a water conservancy hub refers to the water storage area formed by the dam interception; the downstream river section of the spillway refers to the transition area from the outlet of the spillway gate to the flat section of the downstream river channel; the high-sediment content water area refers to the water body with a suspended sediment concentration exceeding 50 kilograms per cubic meter; the underwater robot multi-dimensional scanning data refers to the riverbed topography data collected by an autonomous underwater vehicle equipped with a multi-beam sonar and a three-dimensional laser scanner; the satellite remote sensing full-area coverage data refers to the basin monitoring data fused with high-resolution synthetic aperture radar and multispectral imagery; the sedimentation gradient refers to the rate of change of sediment deposition thickness per unit distance; and the sedimentation spatiotemporal map refers to a three-dimensional model of sedimentation distribution that includes both time and space dimensions.

[0028] In the embodiment of the present application, first, an underwater robot equipped with a 1.5 MHz multi-beam sonar and a 532 nm three-dimensional laser scanner is deployed in the high-sediment-content waters of the water conservancy hub reservoir area and the downstream river section of the spillway. The reservoir area and the downstream river section are scanned with a grid resolution of 0.5 meters to generate riverbed topography point cloud data. For example, the sediment thickness of a certain area scanned is 1.2 meters, the sediment thickness of the adjacent grid is 1.5 meters, and the calculated sediment gradient is 0.3 meters per meter. The full-area coverage data of satellite remote sensing is received synchronously, including satellite multispectral images and synthetic aperture radar data. The satellite data is fused with the underwater scanning data through a point cloud registration algorithm, for example, the 10-meter resolution grid of the satellite image is aligned with the 0.5-meter grid of the underwater scan. Secondly, based on the grid alignment, a spatiotemporal map of silt deposition is constructed based on a three-dimensional interpolation algorithm. Each grid unit in the model contains a sediment gradient parameter. For example, the sediment gradient of a grid unit is 0.05 meters per day. Finally, a filtering algorithm is used to periodically correct the map. The correction period is set to 24 hours, and real-time updates are triggered when the sediment gradient change rate in the key area exceeds 0.01 meters per hour.

[0029] During flood season regulation at a hydropower project in the middle reaches of a river, a swarm of underwater robots equipped with 3D terrain scanning systems was deployed in the reservoir area and the high-sediment-content river section downstream of the spillway. These robots performed high-precision topographic mapping along the longitudinal section of the river. Combined with synthetic aperture radar data from high-resolution satellites, they acquired real-time distribution of suspended sediment concentrations across the entire surface layer. By integrating the underwater robots' topographic point cloud data with satellite-derived sediment dynamics, the system constructed a 3D spatiotemporal map of sediment accumulation, including sedimentation gradients. Every six hours, the system automatically incorporates newly acquired riverbed sediment particle size analysis data to dynamically update the sedimentation rate prediction model in the map, keeping the calculated sediment accumulation error within the reservoir area to within 8%.

[0030] 102. Deploy a radar array in the bend transition section of the spillway outlet. Based on the revised spatiotemporal map of sediment deposition, calculate the corresponding three-dimensional sediment velocity field vector distribution in real time by analyzing the scattering intensity differences and phase shift trajectories of multi-band pulse signals. In step 102, the spillway outlet bend transition section refers to the arc-shaped curved section connecting the spillway and the downstream river channel; the radar array refers to a monitoring system composed of multi-band radars; the multi-band pulse signal scattering intensity difference refers to the difference in echo power after the interaction of electromagnetic waves of different frequencies with sediment particles; the phase offset trajectory refers to the spatial distribution characteristics of the echo signal phase change; and the three-dimensional sediment flow velocity field vector distribution refers to a three-dimensional spatial vector field containing the flow velocity magnitude and direction.

[0031] In this embodiment, a radar array is first deployed in the transition section of the spillway outlet curve, for example, with a spacing of 5 meters between radar array nodes, covering a 120-degree sector-shaped monitoring area. Each radar array node alternately transmits multi-band pulse signals, for example, an X-band pulse signal with a frequency of 9.5 GHz and a Ka-band pulse signal with a frequency of 35 GHz. The receiving end collects data on the scattering intensity differences and phase offset trajectories of the multi-band pulse signals. Secondly, a frequency domain transformation and Doppler shift calculation algorithm are used, for example, with a time resolution of 0.1 seconds, to reconstruct the three-dimensional sediment velocity field vector distribution in real time, generating a velocity vector distribution matrix with a grid accuracy of 0.1 meters per second.

[0032] At the S-shaped bend transition section at the spillway outlet, a multi-band phased array radar monitoring array was deployed along the levees on both sides. Based on a real-time updated spatiotemporal map of sediment deposition, the system transmits frequency-modulated continuous wave (FMCW) signals in different frequency bands, inverts surface water velocity using the differences in scattering intensity of radar echoes, and simultaneously analyzes phase shift trajectories to capture bottom-layer sediment transport characteristics. When the velocity at the bend apex is detected to suddenly increase to 3.5 meters per second, the system uses Doppler effect analysis technology to generate a three-dimensional sediment velocity field vector distribution in real time, accurately identifying the secondary suspension of sediment caused by spiral flow in the outer bend area and providing key fluid dynamics parameters for gate control.

[0033] 103. Dynamically couple the sediment gradient with the three-dimensional sediment velocity field vector distribution to identify the associated region of velocity mutation and sediment deposition within the bend transition section, and generate a constraint condition set including the coordinates of the associated region and a matching rule between velocity and sediment. In step 103, dynamic coupling analysis refers to modeling the correlation between sedimentation and velocity field based on time series; velocity mutation area refers to the area where the local velocity change rate exceeds the threshold; sediment deposition correlation area refers to the overlapping area of ​​velocity mutation and abnormal sedimentation growth; constraint condition set refers to the rule base containing spatial coordinate range, velocity threshold and sedimentation limit.

[0034] In the present embodiment, a dynamic coupling analysis is first performed on the sediment gradient data from the spatiotemporal map of sediment deposition and the three-dimensional sediment velocity field vector distribution, for example, within a 10-minute time window, to calculate a correlation coefficient matrix. When a sudden change in velocity is detected, for example, with a velocity change threshold of 0.5 meters per second per meter and a sediment gradient growth rate exceeding 0.1 centimeters per hour, it is determined to be a sediment deposition correlation region. A machine learning algorithm is then used to generate a set of constraints, for example, defining a sediment deposition correlation region as having a spatial coordinate range of 20 to 30 meters on the X-axis and a maximum allowable sediment growth rate of 0.08 centimeters per hour.

[0035] By dynamically coupling the reservoir's sedimentation gradient model with three-dimensional velocity field data for the bend, the system discovered that when the longitudinal velocity exceeds a specific threshold and the lateral circulation intensity reaches a critical value, a zone of rapid coarse sediment deposition forms at the base of the outer bend revetment. Based on this correlation, the system automatically delineates 12 key coordinate areas and establishes a dynamic matching rule between velocity and sedimentation. When the relationship between the measured velocity and sedimentation in a particular area exceeds a preset threshold, a set of constraints is immediately generated, including spatial coordinates, velocity limits, and sedimentation warning levels, providing a spatialized decision-making basis for gate control.

[0036] 104. Based on the set of constraints, combined with the real-time water level monitoring data and target water level range of the water conservancy hub reservoir area, a dynamic adjustment parameter sequence of the flood discharge gate opening is calculated.

[0037] In step 104, the real-time water level monitoring data refers to the reservoir water level value collected by the sensor array; the target water level range refers to the safe water level range specified in the water conservancy scheduling regulations; and the dynamic adjustment parameter sequence of the flood discharge gate opening refers to a set of gate control instructions sorted by time.

[0038] In an embodiment of the present application, first, real-time water level monitoring data is collected, for example, the sensor accuracy is ±1 cm, and the difference with the target water level range is calculated. Secondly, based on the minimum flow velocity threshold of the sediment deposition-related area in the constraint condition set, for example, 1.2 meters per second, the required flood discharge flow is inverted. Then, the flood discharge flow is calculated using a hydraulic model. For example, if the current inflow flow is 800 cubic meters per second, the flood discharge needs to be increased by 400 cubic meters per second. Finally, a dynamic adjustment parameter sequence for the flood discharge gate opening is generated, for example, the gate opening adjustment step size is 5%, and the time interval is 5 minutes.

[0039] Combining real-time water level monitoring data from the hydropower project's reservoir area with the target water level range, the system autonomously calculates a dynamic control plan for the flood discharge gates based on the set of constraints generated in the previous steps. When the upstream inflow reaches 3,850 cubic meters per second, the system generates phased opening instructions for gates 7 through 9. By dynamically balancing water level fluctuations with the demand for sediment transport through the bend, the reservoir water level is stabilized within a range of plus or minus 0.15 meters of the target value. The system also reduces the average daily sedimentation thickness in the bend transition section from 12 centimeters to 4 centimeters, significantly improving the coordinated control capabilities of reservoir capacity maintenance and flood discharge safety during high-sediment-content floods.

[0040] In summary, through steps 101 to 104, the intelligent closed-loop control technology of sediment transport in the water conservancy hub is realized. The dynamic map of sediment deposition gradient is constructed by the fusion of underwater topography and remote sensing cross-domain data, and the three-dimensional sediment velocity field in the bend area is analyzed by combining multi-band radar scattering inversion technology; based on the spatiotemporal coupling relationship between sediment gradient and velocity vector, the flow-sedimentation synergistic action domain is located, and the adaptive decision model of the flow-sedimentation coupling rule set to drive the flood discharge gate opening is generated, forming a closed-loop optimization chain of sediment transport state perception, multi-parameter dynamic association and gate control strategy, overcoming the spatiotemporal decoupling problem of traditional single-point control, and providing an integrated solution of global dynamic perception and intelligent coordinated control for high-sediment content waters.

[0041] In order to solve the technical problems of difficulty in quantifying and analyzing the dynamic coupling relationship between velocity mutation and sediment deposition in the bend transition section of a river channel and lack of adaptability of the control rules, in some embodiments, the deposition gradient and the three-dimensional sediment velocity field vector distribution are dynamically coupled and analyzed in step 103 to identify the associated area of ​​velocity mutation and sediment deposition in the bend transition section, and generate a set of constraints containing the coordinates of the associated area and the velocity and sediment matching rules, including: 201. Spatially superimpose the time series data of the sediment gradient and the three-dimensional sediment velocity field vector distribution based on the geographic coordinate system of the bend transition section to generate a synchronously updated sediment velocity coupled data set; In step 201, the time series data of the sediment gradient refers to the dataset of sediment thickness change rate recorded in the time dimension; the three-dimensional sediment velocity field vector distribution refers to the three-dimensional space vector field containing the magnitude and direction of the velocity vector; the geographic coordinate system of the bend transition section refers to the local space coordinate system established with the center point of the spillway outlet as the origin; the sediment velocity coupling dataset refers to the fused data generated by spatial superposition that synchronously reflects the dynamic relationship between sedimentation and velocity.

[0042] In an embodiment of the present application, the time series data of the sediment gradient are first processed to unify the spatial coordinates. For example, the UTM projection conversion algorithm is used to convert the sediment gradient data from the geographic coordinate system to a local coordinate system with the center point of the spillway outlet as the origin. The conversion parameters include an easting offset of 500,000 meters, a northing offset of 0 meters, and a central meridian longitude of 120 degrees. Then, spatial resampling is performed on the three-dimensional sediment velocity field vector distribution, and the spatial resolution of the velocity vector is adjusted to a 0.5-meter grid consistent with the sediment gradient data. The data is then superimposed using a bilinear interpolation algorithm. For example, within each grid cell, the interpolation result of the center point is calculated based on the sediment gradient values ​​and velocity vector components of the four adjacent grids to generate a synchronously updated sediment velocity coupled data set. The data synchronization mechanism uses timestamp alignment technology, for example, checking the temporal consistency of sediment and velocity data at a frequency of 10 times per second, and triggering data compensation interpolation when the time deviation exceeds 50 milliseconds.

[0043] 202. In the sediment-flow velocity coupled data set, detect a velocity vector modulus mutation region in the three-dimensional sediment velocity field that meets the sedimentation matching rule, and mark the velocity vector modulus mutation region as a primary velocity mutation region; In step 202, the sedimentation matching rule refers to the empirical correlation model between the velocity vector modulus and the sedimentation gradient; the velocity vector modulus mutation area refers to the local area where the velocity vector length change rate exceeds the set threshold; the primary velocity mutation area refers to the abnormal velocity area marked by the preliminary screening.

[0044] In an embodiment of the present application, a mathematical model of the sedimentation matching rule is first established. For example, the exponential decay relationship between the velocity vector modulus and the sedimentation gradient is obtained by fitting historical data. When the velocity modulus increases by 1 meter per second, the sedimentation gradient decreases by 0.05 centimeters per hour. Then, a sliding window detection method is used to identify the velocity vector modulus mutation area. For example, a 3 by 3 grid unit is used as the detection window to calculate the standard deviation of the velocity modulus in the window. When the standard deviation exceeds the threshold of 0.6 meters per second, it is determined to be a mutation area. Morphological dilation processing is then performed on each mutation area. The structural element is a 3 by 3 square kernel. Discrete noise points are eliminated and adjacent areas are merged. Finally, the processed area is marked as the primary velocity mutation area, a binary mask matrix is ​​generated and stored in association with the sedimentation velocity coupling data set.

[0045] 203. Superimposing the second-order derivative distribution of the sedimentation gradient on the primary velocity mutation area to screen out areas that both meet the extreme value judgment rule and have a sudden change in the sedimentation gradient as velocity-deposition associated areas; In step 203, the second-order derivative distribution of the sedimentation gradient refers to the curvature variation characteristics of the sedimentation gradient in space; the extreme value judgment rule refers to the screening condition of the spatial overlap between the second-order derivative maximum value and the velocity mutation area; the velocity-deposition correlation area refers to the superposition area that satisfies both the velocity mutation and the sedimentation curvature anomaly.

[0046] In the embodiment of the present application, the second-order derivative distribution of the sediment gradient is first calculated. For example, the Sobel operator is used to perform spatial differential operations on the sediment gradient data, and the second-order partial derivatives in the X and Y directions are calculated respectively, and the curvature distribution map is generated by merging. Then, the extreme value judgment rule is set, such as screening the area with a curvature value greater than the threshold of 0.1 cm per square meter, and performing a spatial logical AND operation with the binary mask of the primary velocity mutation area. Then, a connected domain analysis is performed on the overlapping area, and continuous areas with an area greater than 2 square meters are extracted as candidate association areas. Finally, the edges of the candidate areas are smoothed by a boundary optimization algorithm, such as using the Active Contour model to generate a velocity-sediment association area described by a closed polygon, and the coordinates of the regional vertices are stored in a spatial database.

[0047] 204. Extract the three-dimensional geographic coordinate set of the velocity-deposition associated area, and count the numerical correspondence between the sediment gradient and the velocity vector modulus in the velocity-deposition associated area, and generate a constraint condition set including the coordinates of the associated area and the velocity-sediment matching rules based on the numerical correspondence.

[0048] In step 204, the three-dimensional geographic coordinate set refers to the spatial location data set of the associated area; the numerical correspondence refers to the statistical correlation between the sediment gradient and the velocity vector modulus in the area; and the constraint condition set refers to the control rule library including coordinate boundaries and parameter thresholds.

[0049] In an embodiment of the present application, a three-dimensional geographic coordinate set of the velocity-sedimentation correlation area is first derived from a spatial database in the form of a GeoJSON file containing XYZ coordinates and a timestamp. Statistical correlation analysis is then performed, for example, the sedimentation gradient and velocity modulus of all grid cells in each correlation area are extracted, and the Pearson correlation coefficient between the two is calculated. When the correlation coefficient is lower than negative 0.6, it is determined to be a strong negative correlation. A set of constraints is then constructed using a decision tree algorithm, for example, with a velocity modulus of 1.2 meters per second as the splitting threshold, to generate a segmentation rule: when the velocity modulus is not less than 1.5 meters per second, the upper limit of the sedimentation gradient is set to 0.05 centimeters per hour. Finally, the rules are encoded as a set of constraints in JSON format, including regional coordinate boundaries and matching rules for velocity and sedimentation.

[0050] In summary, through steps 201 to 204, a dynamic generation technology of multi-dimensional association rules for sediment transport status is realized. A spatiotemporal coupling model is constructed by spatially synchronously fusing sediment gradient time series data with the three-dimensional velocity field, and primary anomaly areas with sudden changes in velocity vector modulus are automatically detected in the bend geographic coordinate system. Combined with the second-order derivative distribution characteristics of the sediment gradient, a dual-threshold joint criterion of extreme value gradient and velocity mutation is used to screen spatiotemporal overlapping areas, and accurately locate the strong correlation domain between velocity distortion and sedimentation anomaly. Based on the dynamic mapping relationship between the extreme value of sediment gradient and velocity modulus in the correlation domain, a dynamic constraint set containing three-dimensional geographic coordinates and flow-sedimentation parameter matching rules is generated, breaking through the spatiotemporal decoupling defects of traditional single-criteria threshold segmentation, realizing high-confidence identification of sediment transport anomaly areas and adaptive generation of control rules, and providing a multi-parameter coordinated decision-making benchmark for flow state control in water conservancy hubs.

[0051] In order to solve the technical problems of insufficient accuracy in identifying the dynamic coupling area between velocity and sedimentation in the transition section of a river bend and difficulty in associating multi-dimensional spatiotemporal features, in some embodiments, the second-order derivative distribution of the sedimentation gradient is superimposed on the primary velocity mutation area in step 203 to screen out areas that simultaneously meet the extreme value judgment rule and have a sudden sedimentation gradient as velocity-sediment correlation areas, including: 301. Based on the velocity vector length distribution of the spatial unit in the primary velocity mutation zone, set a historical benchmark and a current cycle benchmark for the velocity vector length, and define a primary screening area in the spatial unit that satisfies both the historical benchmark and the current cycle benchmark; In step 301, the primary flow velocity mutation area refers to the abnormal flow velocity area marked by preliminary screening; the historical benchmark refers to the reference value of the flow velocity vector length based on historical data statistics; the current cycle benchmark refers to the statistical reference value of the flow velocity vector length in the current monitoring cycle; the primary screening area refers to a set of spatial units that meet both the historical and current flow velocity benchmarks.

[0052] In an embodiment of the present application, first, based on the distribution of velocity vector lengths of spatial units in the primary velocity mutation zone, a historical benchmark for the velocity vector length is set. For example, the 75th percentile of the velocity modulus over the past 30 days is used as a threshold, and the historical benchmark value is 1.8 meters per second. Then, the current cycle benchmark is calculated. For example, the sliding window average of the current 24-hour velocity modulus is used as a benchmark, and the current benchmark value is 1.5 meters per second. Subsequently, through spatial logical AND operations, spatial units whose velocity vector lengths are higher than both the historical benchmark and the current benchmark are screened out. For example, the velocity modulus in a certain area is 2.1 meters per second, which is higher than the historical benchmark of 1.8 meters per second and the current benchmark of 1.5 meters per second, and the unit is classified into the primary screening area.

[0053] 302. Extract sediment gradient data for each location point in the primary screening area, and superimpose and calculate the second-order derivative distribution of the sediment gradient in adjacent time periods for each location point; In step 302, the sediment gradient data refers to the rate of change of sediment thickness within a unit distance; the adjacent time period refers to a fixed time window before and after the current moment; and the second-order derivative distribution refers to the curvature change characteristics of the sediment gradient in the time dimension.

[0054] In this embodiment, sediment gradient data is first extracted for each location within the primary screening area. For example, the current sediment gradient at coordinate X = 25 meters is 0.12 cm / m. Second-order derivatives within adjacent time periods are then superimposed and calculated for each location. For example, using a 6-hour time window, the central difference method is used to calculate the curvature of the sediment gradient over time. The formula is described as: the current gradient value minus the gradient value for the previous three hours, minus the gradient value for the next three hours minus the current gradient value, and finally divided by the square of the time window length.

[0055] 303. Setting parallel virtual measurement lines in the primary screening area along the water flow direction, generating a discrete point sequence at array intervals for each virtual measurement line, and mapping the second-order derivative distribution calculation results corresponding to the discrete point sequence into a second-order derivative intensity distribution; In step 303, the water flow direction refers to the mainstream flow direction of the river channel downstream of the spillway; the virtual measurement line refers to the manually set monitoring line parallel to the water flow direction; the array spacing refers to the fixed spacing between discrete points on the measurement line; and the second-order derivative intensity distribution refers to the quantitative mapping result of the second-order derivative at the discrete points.

[0056] In the embodiment of the present application, a virtual measurement line is first laid out along the direction of the water flow. For example, five parallel lines spaced 2 meters apart are set within the primary screening area. A sequence of discrete points is then generated based on the array interval. For example, discrete points are generated at 0.5-meter intervals for each measurement line, for a total of 40 points per line. The second-order derivative calculation results corresponding to each discrete point are then mapped to intensity values. For example, a normalization process is used to map the second-order derivative range of 0 to 0.2 cm / h² to an intensity value of 0 to 100, generating an intensity distribution curve along the measurement line.

[0057] 304. Based on the overall distribution state of the second-order derivative intensity distribution, set an extreme value determination rule, and mark adjacent discrete point areas that continuously meet the extreme value determination rule as sedimentation gradient mutation areas; In step 304, the extreme value judgment rule refers to the abnormal area screening condition set based on the second-order derivative intensity distribution; continuous satisfaction means that the intensity values ​​of adjacent discrete points continue to exceed the threshold; the sedimentation gradient mutation area refers to the continuous spatial area that meets the extreme value judgment.

[0058] In this embodiment, an extreme value determination rule is first set based on the overall distribution of the second-order derivative intensity distribution. For example, an intensity value exceeding 80 and at least three consecutive discrete points satisfying the condition are required. Next, a sequence of discrete points is scanned along a virtual measurement line. For example, on measurement line L1, the intensity values ​​of points 10 to 15 are detected to be 85, 89, 92, 88, 84, and 81, respectively, and these points are identified as consecutive regions satisfying the condition. Finally, adjacent mutation regions are merged using a morphological closing operation. For example, regions separated by less than 1 meter are merged into a single sedimentation gradient mutation region.

[0059] 305. The sediment gradient mutation zone is superimposed in three-dimensional space along the vertical direction of the virtual measurement line, and the spatial coordinates of the primary screening area and the distribution range of the sediment gradient mutation zone are integrated. The set of superimposed spatial units that simultaneously meet the velocity vector length benchmark condition and the extreme value judgment rule is defined as the velocity deposition association area.

[0060] In step 305, three-dimensional spatial superposition refers to the lateral data fusion along the direction perpendicular to the water flow; the velocity vector length reference condition refers to the dual velocity threshold defined in step 301; the velocity-sedimentation association area refers to the set of spatial units that simultaneously meet the velocity and deposition anomaly conditions.

[0061] In this embodiment, the sediment gradient mutation zones on each virtual measurement line are first spatially interpolated perpendicular to the water flow direction, for example, using the inverse distance weighted method to generate a horizontally continuous three-dimensional mutation zone distribution. A logical AND operation is then performed on the spatial coordinates of the primary screening area and the mutation zone distribution. For example, an area within the coordinate range of X = 24-26 meters and Y = 3-5 meters is selected where both the velocity modulus ≥ 1.5 m / s and the sediment gradient mutation intensity ≥ 80 are satisfied. The superposition result is ultimately defined as a velocity-sedimentation correlation region, and its boundary coordinate set is output.

[0062] In summary, steps 301 to 305 demonstrate a multi-reference spatiotemporal joint location technique for sediment transport-related regions. This technique constructs primary anomaly spatial units based on historical and current dual-period baselines of velocity vector length. Incorporating the spatiotemporal evolution of the second-order derivative of the sediment gradient, a gridded discrete analysis system of virtual measurement lines is deployed along the flow direction. Dynamic threshold criteria are used to automatically calibrate the distribution zones of sediment gradient extremes. A three-dimensional spatial superposition and fusion algorithm of velocity references and gradient extremes is employed to achieve multidimensional parameter coupling verification of primary velocity anomaly zones and sediment mutation zones, accurately identifying complex association regions that meet both velocity vector length constraints and second-order derivative extreme value conditions. This method overcomes the limitations of single-dimensional static threshold determination and enhances the robustness of identifying flow-sedimentation association regions through a multi-reference spatiotemporal joint analysis mechanism. This provides a decision-making basis for sediment transport control at water conservancy hubs that combines spatiotemporal continuity with parameter coordination.

[0063] In order to solve the technical problems of insufficient fusion of vertical and surface data and poor real-time performance in the dynamic analysis of three-dimensional sediment velocity field, in some embodiments, the step 102, based on the modified sediment deposition spatiotemporal map, calculates the corresponding three-dimensional sediment velocity field vector distribution in real time through the scattering intensity difference and phase offset trajectory of the multi-band pulse signal, including: 401. Synchronously transmit a multi-band pulse signal including a low-frequency penetration signal and a high-frequency tracking signal, and after receiving the reflected signal, separate and generate a sediment density distribution map penetrating to the bottom of the riverbed and a sediment movement trajectory map on the surface; In step 401, the low-frequency penetrating signal refers to a long-wavelength electromagnetic wave or sound wave signal that can penetrate to the bottom of the riverbed; the high-frequency tracking signal refers to a short-wavelength signal used to capture the movement of surface sediment particles; the sediment density distribution map refers to tomographic imaging data reflecting the sediment accumulation density in the vertical direction of the riverbed; the surface sediment movement trajectory map refers to the surface particle displacement trajectory reconstructed by the high-frequency signal reflection characteristics.

[0064] In this embodiment, multi-band pulse signals are first transmitted synchronously. For example, the low-frequency signal uses a 10 kHz sound wave with a penetration depth of 5 meters to generate riverbed density data. The high-frequency signal uses a 100 MHz radar wave with a resolution of 0.1 meters to track sediment movement within the surface layer within 0.5 meters. After receiving the reflected signal, the low-frequency and high-frequency components are separated through frequency domain filtering. For example, a Fourier transform is used to extract the low-frequency band from 10 kHz to 100 kHz to generate a sediment density distribution map. The high-frequency band from 1 MHz to 100 MHz is extracted and a feature point matching algorithm is used to generate a surface sediment movement trajectory map.

[0065] 402. Extract the sediment accumulation density change rate of each depth layer in the vertical direction from the sediment density distribution map, and combine it with the sediment amount gradient at the corresponding position in the revised sediment deposition spatiotemporal map to generate a vertical density gradient correlation profile; In step 402, the sediment accumulation density change rate of each depth layer in the vertical direction refers to the increase or decrease rate of the density value within a unit depth; the vertical density gradient correlation profile refers to a vertical section model that spatially correlates the density change rate with the sedimentation gradient.

[0066] In the embodiment of the present application, the vertical density change rate is first extracted from the sediment density distribution map. For example, the density is 1.8 grams per cubic centimeter at a depth of 2 meters, and the density is 2.1 grams per cubic centimeter at a depth of 2.5 meters. The calculated change rate is 0.6 grams per cubic centimeter per meter. Then, the sedimentation gradient at the same location is obtained from the corrected sediment deposition spatiotemporal map. For example, the sedimentation gradient at this location is 0.15 centimeters per meter. The two are superimposed by a spatial interpolation algorithm. For example, in the depth range of 2 to 3 meters, a correlation profile is generated with the density change rate as the horizontal axis and the sedimentation gradient as the vertical axis, and the density gradient mutation area is marked, such as the area where the change rate exceeds 0.5 grams per cubic centimeter per meter.

[0067] 403. Perform three-dimensional spatial decomposition on the surface sediment movement trajectory map, and establish a surface particle movement direction angle distribution map based on the spatial position offset of the reflection signal feature points within the continuous transmission cycle; In step 403, three-dimensional spatial decomposition refers to expanding the two-dimensional surface motion trajectory data into three-dimensional space; the reflection signal feature point refers to the high-frequency echo scattering center of the surface sediment particles; the spatial position offset refers to the coordinate change of the feature point within the continuous signal period; the surface particle motion direction angle distribution map refers to the spatial distribution model reflecting the deflection angle of the surface particle motion direction.

[0068] In an embodiment of the present application, the surface sediment movement trajectory map is first subjected to three-dimensional decomposition, for example, the two-dimensional plane coordinates X, Y are superimposed with the water depth data Z to reconstruct the three-dimensional movement path of the particles. Then, the offset of the feature points within the continuous emission cycle is extracted. For example, the coordinates of a feature point at time t1, such as 10 meters, 5 meters, 0.2 meters, move to t2, such as 10.3 meters, 5.1 meters, 0.2 meters, and the displacement vector is calculated, such as ΔX=0.3 meters, ΔY=0.1 meters. Then, through the vector direction angle calculation formula, such as the direction angle θ=arctan(ΔY / ΔX), the movement direction angle of the point is obtained to be 18.4 degrees. Finally, all feature points are counted to generate a direction angle distribution map with 5-degree intervals.

[0069] 404. Spatially match the vertical density gradient correlation profile with the surface particle movement direction angle distribution map, superimpose the surface movement direction angle data of the corresponding position in the vertical density gradient mutation area, and generate the corresponding three-dimensional sediment velocity field vector distribution through vector direction transfer.

[0070] In step 404, spatial position matching refers to aligning the geographic coordinates of the vertical profile data with the surface directional angle data; the vertical density gradient mutation zone refers to the depth interval where the density change rate exceeds the threshold; and vector direction transfer refers to an algorithm for inferring the bottom flow velocity vector based on the surface movement direction.

[0071] In an embodiment of the present application, first, coordinate matching is performed between the vertical density gradient correlation profile and the surface directional angle distribution map. For example, at the plane coordinates X=20 meters, Y=10 meters, the vertical profile covers a depth of 0 to 5 meters, and the surface directional angle data corresponds to a water depth of 0 to 0.5 meters. Then, the surface directional angle is superimposed in the density gradient mutation area, such as a depth of 2-3 meters. For example, the average surface directional angle in this area is 20 degrees. Through the direction transfer algorithm, it is assumed that there is a linear correlation between the bottom flow velocity direction and the surface layer, for example, the bottom direction angle = the surface direction angle × 0.8, and the flow velocity direction at a depth of 2 meters is generated to be 16 degrees. Finally, the direction and density gradient data of each depth layer are fused to generate a three-dimensional sediment velocity field vector distribution.

[0072] In summary, steps 401 to 404 implement a multi-band collaborative inversion technique for riverbed sediment motion. This technique uses a composite detection mechanism combining low-frequency penetrating signals and high-frequency tracking signals to simultaneously acquire the vertical density distribution of the riverbed and the trajectories of surface particle motion. A vertical density-sedimentation correlation model is constructed based on the spatial mapping of the vertical density gradient change rate and the sediment gradient. Simultaneously, a three-dimensional decomposition technique for motion trajectories is used to invert the angular distribution of surface particle motion. A vertical-to-surface motion vector transfer function is established by spatially matching vertical density mutation zones with surface directional angles. Multi-source data is then integrated to generate a three-dimensional dynamic distribution model of the sediment velocity field. This method overcomes the sensing limitations of traditional single-dimensional detection, enabling cross-scale dynamic correlation analysis of the evolution of riverbed vertical structure and surface flow patterns. This approach ensures the accuracy of three-dimensional vector inversion under the constraints of vertical density gradients for velocity field reconstruction in highly sediment-laden waters.

[0073] In order to solve the problem of missing fusion of vertical and horizontal motion data and distortion of vector direction transfer in the reconstruction of three-dimensional sediment velocity field, in some embodiments, step 404 includes spatially matching the vertical density gradient correlation profile with the surface particle motion direction angle distribution map, superimposing the surface motion direction angle data at the corresponding position in the vertical density gradient mutation area, and generating the corresponding three-dimensional sediment velocity field vector distribution through vector direction transfer, including: 501. Spatially matching the data of each depth layer of the vertical density gradient correlation profile with the plane coordinates of the surface particle movement direction angle distribution map to screen out the vertical density gradient mutation area corresponding to the plane coordinates; In step 501, the vertical density gradient correlation profile refers to the tomographic data reflecting the correlation between the density change in the vertical direction of the riverbed and the sediment gradient; the surface particle movement direction angle distribution map refers to the planar distribution model of the deflection angle of the surface sediment particle movement direction; the vertical density gradient mutation zone refers to the vertical depth interval where the density change rate exceeds the set threshold; the plane coordinates refer to the X-axis and Y-axis positions in the two-dimensional geographic coordinate system.

[0074] In the present embodiment, the spatial position of each depth layer data of the vertical density gradient associated profile is first matched with the plane coordinates of the surface particle movement angular distribution map. For example, a geo-registration algorithm is used to align the grid data of the vertical profile with the grid coordinate system of the surface angular data. Next, the vertical density gradient mutation area corresponding to the plane coordinates is screened. For example, at coordinates X equal to 20 meters and Y equal to 5 meters, the vertical profile shows that the density change rate from a depth of 2 meters to 3 meters reaches 0.8 grams per cubic centimeter per meter, exceeding the threshold of 0.5 grams per cubic centimeter per meter. This area is marked as a vertical density gradient mutation area.

[0075] 502. Extracting surface motion direction angle data of the plane coordinate corresponding to the vertical density gradient mutation area, and calculating the horizontal motion vector of each plane coordinate point based on the surface motion direction angle data; In step 502, the surface movement direction angle data refers to the horizontal movement direction angle of the surface sediment particles at the plane coordinate point; the horizontal movement vector refers to the two-dimensional plane vector synthesized by the direction angle and the velocity modulus; the plane coordinate point refers to the specific position of the X-axis and the Y-axis in the two-dimensional geographic coordinate system.

[0076] In this embodiment, surface motion azimuth data for the plane coordinates corresponding to the region of vertical density gradient abrupt change is first extracted. For example, the surface azimuth angle for coordinates 20 meters (X) and 5 meters (Y) is 25 degrees. The horizontal motion vector for each plane coordinate point is then calculated based on this surface motion azimuth data. For example, if the surface velocity modulus is 1.2 meters per second, the horizontal vector's X component is 1.2 times the cosine of 25 degrees, and its Y component is 1.2 times the sine of 25 degrees, resulting in a horizontal motion vector with an X component of 1.09 meters per second and a Y component of 0.51 meters per second.

[0077] 503. Determine the gradient direction change amount of each position point in the vertical density gradient mutation region, and calculate the corresponding vertical direction correction value based on the gradient direction change amount; In step 503, the gradient direction change refers to the deflection angle of the density change direction of different depth layers in the vertical density gradient mutation area; the vertical direction correction value refers to the adjustment value of the vertical component of the vector based on the gradient direction change; the position point refers to the X-axis, Y-axis and Z-axis coordinate points in three-dimensional space.

[0078] In this embodiment, the gradient direction change at each point in the vertical density gradient abrupt change region is first determined. For example, at a depth of 2 meters, the density gradient direction is deflected downward by 30 degrees, while at a depth of 3 meters, it becomes deflected downward by 45 degrees, resulting in a calculated gradient direction change of 15 degrees. Based on this gradient direction change, the corresponding vertical correction value is then calculated. For example, a 5-degree change per meter of depth corresponds to a vertical component correction factor of 0.1 meters per second, resulting in a vertical correction value of 0.3 meters per second from a depth of 2 to 3 meters.

[0079] 504. After superimposing the vertical correction value on the horizontal motion vector, the vector is transferred layer by layer along the vertical density gradient change direction, and the vector direction is corrected based on the offset angle between the gradient direction and the surface direction angle. In step 504, the offset angle between the gradient direction and the surface direction angle refers to the angle between the vertical density gradient direction and the surface motion direction; the vector direction correction refers to adjusting the three-dimensional direction of the motion vector according to the offset angle; the gradient direction change direction refers to the deflection trend of the vertical density gradient with increasing depth.

[0080] In the embodiment of the present application, the horizontal motion vector is first superimposed with the vertical correction value. For example, at coordinates X equal to 20 meters and Y equal to 5 meters, the horizontal vector has an X component of 1.09 meters per second and a Y component of 0.51 meters per second. A vertical correction of 0.3 meters per second is superimposed to generate an initial three-dimensional vector with an X component of 1.09 meters per second, a Y component of 0.51 meters per second, and a Z component of 0.3 meters per second. Next, the offset angle between the gradient direction and the surface direction angle is calculated. For example, a horizontal offset angle of 10 degrees from a 25-degree surface direction to the gradient direction is 35 degrees. The vector direction is corrected based on the offset angle. For example, the three-dimensional vector is rotated 10 degrees about the Z axis to obtain a corrected vector with an X component of 1.05 meters per second, a Y component of 0.61 meters per second, and a Z component of 0.3 meters per second.

[0081] 505. Spatially superimpose the corrected motion vectors of all depth layers in the vertical density gradient mutation area to generate a three-dimensional sediment velocity field vector distribution covering corresponding plane coordinates.

[0082] In step 505, the corrected motion vector refers to the three-dimensional velocity vector that has been direction-corrected; the coverage of the corresponding plane coordinates refers to the X-axis and Y-axis ranges of the target area covered by the three-dimensional spatial model; and the three-dimensional sediment velocity field vector distribution refers to the three-dimensional continuous vector field containing the velocity magnitude and direction.

[0083] In this embodiment, the corrected motion vectors for all depths within the vertical density gradient abrupt change are first spatially superimposed. For example, linear interpolation is performed on the corrected vectors at depths of 2 meters, 2.5 meters, and 3 meters. A three-dimensional sediment velocity field vector distribution is then generated, covering the corresponding plane coordinates. For example, within the plane coordinate range of 18 to 22 meters for X and 4 to 6 meters for Y, the three-dimensional velocity vector for each grid cell is output.

[0084] In summary, steps 501 to 505 demonstrate a cross-dimensional dynamic vector reconstruction technique for riverbed sediment velocity fields. Vertical mutation zones are located by spatially matching vertical density gradient correlation profiles with the surface motion angular distribution. Based on the multidirectional coupled calculation of surface horizontal motion vectors and vertical gradient corrections, a gradient-driven vector dynamic transfer algorithm is employed to recursively extend the horizontal motion component vertically. Spatial directional correction of the motion vector is achieved through an offset angle compensation mechanism between the vertical gradient change direction and the surface directional angle. Finally, the correction vectors from each depth layer are integrated to construct a three-dimensional sediment velocity field distribution model. This method overcomes the dimensional limitations of traditional two-dimensional flow field extrapolation, achieving a cross-scale dynamic correlation between vertical density gradient mutations and surface motion vectors. This provides the capability for accurately reconstructing three-dimensional velocity fields under vertical structural constraints in highly sediment-rich waters.

[0085] In order to solve the problems of insufficient multi-source data fusion and delayed map correction in dynamic monitoring of riverbed sedimentation, in some embodiments, the step 101 includes synchronously acquiring multi-dimensional scanning data of the riverbed topography by an underwater robot and global coverage data of satellite remote sensing to generate a spatiotemporal map of sedimentation including a sediment gradient, and periodically correcting the spatiotemporal map of sedimentation based on changes in the sediment gradient, including: 601. Set the underwater robot's scanning path for the riverbed topography, synchronously obtain sediment accumulation thickness and multi-dimensional scanning data at each scanning point, and synchronously obtain full-area coverage data from satellite remote sensing during the robot's scanning intervals; In step 601, the underwater robot scanning path refers to the pre-planned riverbed terrain detection trajectory; the sediment accumulation thickness refers to the vertical height of the sediment from the riverbed surface to the stable layer below; the multi-dimensional scanning data refers to the comprehensive detection data including sonar reflection intensity, terrain undulation and bottom hardness; the scanning interval period refers to the moving waiting time after the robot completes a single scan; the satellite remote sensing full-area coverage data refers to the surface coverage information of the watershed area obtained by optical or radar satellites.

[0086] In an embodiment of the present application, the scanning path of the underwater robot is first set, for example, a spiral path is used to cover the reservoir area at a spacing of 20 meters, and each scanning point is stopped for 10 minutes to obtain sediment accumulation thickness data. For example, the thickness of a certain point is measured to be 2.3 meters. The sediment accumulation thickness and multi-dimensional scanning data of each scanning point are obtained simultaneously, for example, the sonar reflection intensity value is -30 decibels and the bottom hardness is medium sand grade. During the interval period when the robot moves to the next scanning point, the satellite remote sensing full-area coverage data is received during the robot scanning interval period, such as obtaining the multispectral image of the Sentinel-2 satellite and the synthetic aperture radar data of the Sentinel-1 satellite.

[0087] 602. Calculate the thickness variation difference of the sediment accumulation thickness at each scanning point according to the continuous scanning period, use the thickness variation difference and the ratio of the horizontal spacing between adjacent scanning points as the sediment amount gradient, and combine the multi-dimensional scanning data and the global coverage data to generate a spatiotemporal map of sediment deposition including the gradient direction and sediment amount gradient; In step 602, the thickness change difference refers to the difference in sediment thickness between adjacent scanning cycles at the same scanning point; the horizontal spacing ratio refers to the straight-line distance between two adjacent scanning points; the sedimentation gradient refers to the change in sediment thickness per unit horizontal distance; the gradient direction refers to the spatial direction of the maximum change rate of the sedimentation gradient; and the spatiotemporal map of sedimentation refers to a sedimentation dynamic model that integrates time, space, and multi-dimensional attributes.

[0088] In an embodiment of the present application, the sediment accumulation thickness of the scanning point is first calculated according to the change difference of continuous cycles. For example, the thickness of the previous cycle is 2.3 meters, the current cycle is 2.5 meters, and the thickness change difference is 0.2 meters. Then the horizontal spacing ratio of adjacent scanning points is calculated, and the thickness change difference and the horizontal spacing ratio of adjacent scanning points are used as the sedimentation gradient. For example, the distance between point A and point B is 10 meters, and the sedimentation gradient is 0.2 meters divided by 10 meters, which is equal to 0.02 meters per meter. Combining multi-dimensional scanning data and global coverage data, for example, the sonar reflection intensity of point A is -30 decibels, corresponding to the bottom type of fine sand, and satellite images show that the vegetation coverage in this area is less than 5%, and finally a spatiotemporal map of sediment deposition with a gradient direction of 15 degrees southeast is generated.

[0089] 603. When the direction of the deposition gradient at any scanning point continuously deviates from the initial direction by more than a set angle, or the fluctuation amplitude of the deposition gradient value exceeds a set ratio, the scanning path encryption of the area where the point is located is triggered; In step 603, the initial direction refers to the spatial orientation of the deposition gradient during the first scan; the set angle refers to the maximum threshold value allowed for the gradient direction to deviate; the set ratio refers to the maximum allowable range of fluctuations in the deposition gradient value; and the scanning path encryption refers to increasing the density of scanning points in abnormal areas.

[0090] In this embodiment, the gradient direction of the scanning point is first monitored. For example, if the initial direction is 15 degrees east-southeast, the direction is detected to have shifted to 30 degrees due east in the current cycle. The deviation angle reaches 15 degrees, exceeding the set threshold of 10 degrees. Alternatively, a fluctuation in the sediment gradient value is detected, such as a sudden increase from 0.02 meters per meter to 0.03 meters per meter, with a fluctuation amplitude of 50% exceeding the set ratio of 30%. When any of these conditions are met, the path encryption for that area is triggered, for example, shortening the original 20-meter scanning interval to 5 meters and adding a circular supplementary scanning path in the area.

[0091] 604. Re-acquire the sediment accumulation thickness in the encrypted scanning area, and modify the value and direction of the sediment gradient based on the new sediment accumulation thickness data, overwriting the data of the corresponding area in the original spatiotemporal map.

[0092] In step 604, the encrypted scanning area refers to the high-density detection range added after the path adjustment; the modified sedimentation gradient refers to updating the original gradient parameters based on the new data; and the data coverage refers to replacing the corresponding part of the original spatiotemporal map with the new calculation results.

[0093] In this embodiment of the present application, the sediment accumulation thickness is first re-acquired in the encrypted scanning area. For example, three sub-points are added to the original abnormal point, and the measured thicknesses are 2.6 meters, 2.55 meters, and 2.58 meters, respectively. The sediment gradient is calculated based on the new data. For example, the thickness change difference of 5 meters horizontally is 0.05 meters, the gradient is corrected to 0.01 meters per meter, and the direction is adjusted to 10 degrees southeast. Finally, the corrected gradient value and direction are overwritten with the original space-time map, for example, replacing the old data in the coordinate range of X = 100-105 meters and Y = 50-55 meters.

[0094] In summary, steps 601 to 604 demonstrate an adaptive dynamic optimization technique for riverbed sedimentation monitoring. By leveraging the collaborative acquisition of global data from satellite remote sensing and multi-dimensional scanning by an underwater robot, a spatiotemporal evolution analysis model driven by sediment gradients is constructed. Based on the difference in sediment thickness changes and spatial gradient calculations within consecutive scanning cycles, a spatiotemporal map containing gradient direction and intensity information is dynamically generated. A dual-modal anomaly detection mechanism based on gradient direction deviation and fluctuation amplitude is introduced to trigger an autonomous scanning path encryption strategy. Local rescanning data is then used to correct gradient parameters and the spatial distribution of the map in real time. This method overcomes the response hysteresis inherent in fixed-path scanning and achieves a balanced optimization of monitoring accuracy and resource efficiency through a dynamic feedback mechanism triggered by gradient anomalies, forming a closed loop of dynamic riverbed topography perception characterized by "global coverage, anomaly localization, and local enhancement."

[0095] In order to solve the technical problems of insufficient dynamic coupling of water level, flow velocity, and sedimentation parameters and insufficient safety of adjustment parameters in reservoir flood discharge gate regulation, in some embodiments, the dynamic adjustment parameter sequence of the flood discharge gate opening is calculated based on the constraint condition set, combined with the real-time water level monitoring data of the water conservancy hub reservoir area and the target water level range, in step 104, including: 701. Obtain real-time water level monitoring data for the reservoir area of ​​the water conservancy hub, calculate the deviation between the current water level and the median of the target water level interval, and predict the intensity of water level change trends in future periods based on historical water level change data; In step 701, the real-time water level monitoring data refers to the current water level value of the reservoir area collected by the pressure sensor array; the median of the target water level interval refers to the midpoint value of the safe water level range specified in the water conservancy scheduling regulations; the water level deviation refers to the algebraic difference between the current water level and the target median; the water level change historical data refers to the water level fluctuation record within the past set time period; the water level change trend intensity in the future period refers to the quantitative indicator of the water level rise and fall rate predicted based on historical data.

[0096] In an embodiment of the present application, data is first collected through a real-time water level monitoring network of six groups of pressure sensors deployed in the reservoir area. For example, the sensor nodes upload water level values ​​at a sampling frequency of once per second. The system performs redundancy checking and mean filtering on the six groups of data, removes outliers, and takes the arithmetic mean as the current water level value. If the current water level is 100.6 meters above sea level, the median of the target water level interval is 100.75 meters, and the calculated water level deviation is negative 0.15 meters. Then, the water level data for the past three hours is extracted from the historical database. For example, the hourly records are 100.70 meters, 100.65 meters, and 100.62 meters. The water level change rate is calculated to be 0.03 meters per hour by the first-order difference method, and the trend intensity for the next two hours is predicted to be 0.025 meters per hour based on the autoregressive sliding average model. The prediction result is marked with a quantitative level of "high".

[0097] 702. Analyze the flow velocity threshold and the sedimentation gradient threshold of each associated area in the constraint condition set, convert the ratio of the current measured flow velocity to the flow velocity threshold into a flow velocity adjustment parameter, and convert the ratio of the current sedimentation gradient to the sedimentation gradient threshold into a sedimentation gradient adjustment parameter. In step 702, the constraint condition set refers to a rule base including the three-dimensional geographic coordinates of the associated area, the flow velocity threshold and the sedimentation gradient threshold; the flow velocity adjustment parameter refers to the ratio of the current measured flow velocity to the flow velocity threshold; and the sedimentation gradient adjustment parameter refers to the ratio of the current sedimentation gradient to the sedimentation gradient threshold.

[0098] In an embodiment of the present application, the rule parameters of the associated area R001 are first extracted from the constraint condition set. For example, the regional coordinate range is X equal to 20 to 25 meters, Y equal to 5 to 10 meters, and Z equal to 0 to 5 meters, the flow velocity threshold is 1.5 meters per second, and the sedimentation gradient threshold is 0.08 centimeters per hour. The current measured flow velocity is 1.2 meters per second obtained by the underwater acoustic Doppler current profiler deployed in the area, and the current sedimentation gradient is 0.1 centimeters per hour extracted from the spatiotemporal map of silt deposition. When calculating the flow velocity adjustment parameter, the measured flow velocity of 1.2 meters per second is divided by the flow velocity threshold of 1.5 meters per second to obtain a ratio of 0.8; the sedimentation gradient adjustment parameter is the measured gradient of 0.1 centimeters per hour divided by the threshold of 0.08 centimeters per hour to obtain a ratio of 1.25. If the measured value exceeds the threshold, the parameter value is marked as a red warning state.

[0099] 703. Perform weighted fusion on the flow velocity adjustment parameter and the sedimentation gradient adjustment parameter to generate initial adjustment parameter values ​​for each associated area, and apply directional correction to the initial adjustment parameter values ​​based on the water level deviation direction and trend strength; In step 703, weighted fusion refers to the linear combination of the flow rate adjustment parameter and the sedimentation gradient adjustment parameter according to preset weights; the initial adjustment parameter value refers to the comprehensive adjustment coefficient after fusion; and the directional correction refers to the adjustment of the sign or amplitude of the parameter value according to the direction of the water level deviation and the trend strength.

[0100] In the embodiment of the present application, the weight coefficients are first set according to the water conservancy scheduling strategy: the flow rate adjustment parameter weight is 0.6, and the sedimentation gradient adjustment parameter weight is 0.4. The adjustment parameters of area R001 are weighted and fused, and the initial adjustment parameter value is calculated to be 0.8 times 0.6 plus 1.25 times 0.4, which is 0.98. The parameter sign is then corrected according to the direction of the water level deviation: because the current water level is lower than the target median and the predicted trend is a decline, the flood discharge needs to be reduced, so the parameter value of 0.98 is multiplied by negative 1 to obtain negative 0.98. At the same time, based on the trend intensity level "high", a gain is applied to the parameter amplitude, for example, negative 0.98 is multiplied by 1.2 times, and the final corrected parameter value is negative 1.176. This value indicates that the flood discharge needs to be reduced to below the baseline value of 1.176 times the current flow.

[0101] 704. Arrange the corrected adjustment parameter values ​​of each area in a time series to ensure that the parameter variation in adjacent time periods does not exceed the gate mechanical action safety threshold, and smooth the parameter differences between spatially adjacent areas to generate a dynamic adjustment parameter sequence for the flood discharge gate opening.

[0102] In step 704, time series arrangement refers to arranging parameter values ​​in chronological order; the mechanical action safety threshold refers to the maximum allowable range of a single adjustment of the gate opening; the spatially adjacent area refers to the associated area with adjacent geographic coordinates; and smoothing processing refers to reducing parameter mutations through interpolation or filtering algorithms.

[0103] In an embodiment of the present application, the corrected parameter values ​​of each area are first divided into time windows, for example, the next two hours are divided into 8 15-minute time periods, and an initial sequence is generated: negative 1.176, negative 1.128, negative 1.080, negative 1.032, negative 0.984, etc. Then the amplitude of the parameter change in adjacent time periods is checked. For example, the amplitude change from negative 1.176 in the first time period to negative 1.128 in the second time period is 4.1%, and the amplitude change from the second time period to the third time period is 4.2%, which are all within the allowable range. If it is detected that the parameter jump in a certain time period exceeds the threshold, for example, jumping directly from negative 1.176 to negative 1.032, the intermediate parameters of negative 1.104 and negative 1.068 are inserted for a smooth transition. Finally, a Gaussian filter is applied to the parameter differences between R001 and R002 in spatially adjacent regions. If the R001 parameter is negative 1.080 and R002 is negative 0.950, representing a 13% difference, then a weighted average is used to adjust R001 to negative 1.015 and R002 to negative 0.985, reducing the difference to 3%. The final output parameter sequence is sent through the control loop to the gate actuator, driving the hydraulic system to adjust the opening in steps.

[0104] In summary, steps 701 to 704 demonstrate a multimodal adaptive decision-making technique for flood discharge gates. By dynamically sensing real-time water level deviations and trend strength, combined with a dual-threshold parameter analysis mechanism for flow-sedimentation regions, a multi-objective collaborative optimization model for flow velocity and sedimentation gradient adjustment parameters is constructed. A directional correction algorithm driven by water level conditions is used to dynamically calibrate the optimization parameters. A spatiotemporal smoothing constraint mechanism ensures the mechanical safety of the gate control sequence and the coordination of regional flow patterns, forming a closed-loop control strategy that integrates multiple factors: water level, flow pattern, and sedimentation. This method breaks through the rigid constraints of traditional single-threshold control, achieving dynamic optimization of gate opening within a safety boundary and collaborative adaptation of multi-regional parameters. This provides intelligent decision-making support for both flood control and sedimentation prevention in highly sediment-rich waters.

[0105] Figure 2 A schematic diagram of a water level regulation system for a water conservancy project is provided for the present invention. Figure 2 As shown, the system includes: Acquisition module 21, for the high sediment concentration waters in the reservoir area of ​​the water conservancy hub and the downstream river section of the spillway, synchronously acquires multi-dimensional scanning data of the riverbed topography by underwater robots and global coverage data of satellite remote sensing, so as to generate a spatiotemporal map of sediment deposition including a sediment gradient, and periodically amends the spatiotemporal map of sediment deposition based on changes in the sediment gradient; A calculation module 22 deploys a radar array at the curve transition section of the spillway outlet, and calculates the corresponding three-dimensional sediment velocity field vector distribution in real time based on the modified spatiotemporal map of sediment deposition by scattering intensity differences and phase shift trajectories of multi-band pulse signals; A generation module 23 performs a dynamic coupling analysis on the sediment gradient and the three-dimensional sediment velocity field vector distribution, identifies the associated region of velocity mutation and sediment deposition in the bend transition section, and generates a constraint condition set including the coordinates of the associated region and the flow velocity and sediment matching rule; The calculation module 24 calculates a dynamic adjustment parameter sequence of the flood discharge gate opening according to the constraint condition set and in combination with the real-time water level monitoring data and the target water level range of the water conservancy hub reservoir area.

[0106] Figure 2 The water level regulating system of a water conservancy hub can be implemented Figure 1 The implementation principle and technical effects of the water level regulation method for a water conservancy hub described in the illustrated embodiment will not be elaborated on here. The specific manner in which each module and unit performs operations in the water level regulation system for a water conservancy hub in the above embodiment has been described in detail in the embodiment of the method and will not be elaborated on here.

[0107] In one possible design, Figure 2 The water level regulation system of a water conservancy hub in the embodiment shown can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32; The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0108] The processing component 32 is used for the above Figure 1 The embodiment provides a method for regulating the water level of a water conservancy project.

[0109] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0110] The storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0111] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0112] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0113] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0114] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0115] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The embodiment shown is a method for regulating the water level of a water conservancy project.

[0116] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0117] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0118] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0119] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for regulating the water level of a water conservancy hub, characterized in that: include: For high-sediment-content waters in the reservoir area of ​​the water conservancy hub and the downstream section of the spillway, multi-dimensional scanning data of the riverbed topography by underwater robots and global coverage data from satellite remote sensing are simultaneously obtained to generate a spatiotemporal map of sediment deposition that includes a sediment gradient. This spatiotemporal map is then periodically revised based on changes in the sediment gradient. A radar array is deployed in the bend transition section of the spillway outlet. Based on the modified spatiotemporal map of sediment deposition, the corresponding three-dimensional sediment velocity field vector distribution is calculated in real time by using the scattering intensity differences and phase shift trajectories of multi-band pulse signals. Dynamically coupling the sediment gradient with the three-dimensional sediment velocity field vector distribution to identify the associated region of velocity mutation and sediment deposition within the bend transition section, and generating a constraint condition set including the coordinates of the associated region and the flow velocity and sediment matching rule; According to the set of constraints, combined with the real-time water level monitoring data and the target water level range of the water conservancy hub reservoir area, a dynamic adjustment parameter sequence of the flood discharge gate opening is calculated.

2. The method according to claim 1, characterized in that Dynamically couple the sediment gradient with the three-dimensional sediment velocity field vector distribution to identify the associated region of velocity mutation and sediment deposition within the bend transition section, and generate a set of constraints containing the coordinates of the associated region and the velocity and sediment matching rules, including: Spatially superimposing the time series data of the sediment gradient and the three-dimensional sediment velocity field vector distribution based on the geographic coordinate system of the bend transition section to generate a synchronously updated sediment velocity coupled data set; In the sediment-flow velocity coupled data set, detecting a velocity vector modulus mutation region in the three-dimensional sediment velocity field that meets the sedimentation matching rule, and marking the velocity vector modulus mutation region as a primary velocity mutation region; superimposing the second-order derivative distribution of the sedimentation gradient on the primary velocity mutation area to screen out areas that simultaneously meet the extreme value judgment rule and where the sedimentation gradient undergoes a mutation as velocity-deposition associated areas; The three-dimensional geographic coordinate set of the velocity-deposition associated area is extracted, and the numerical correspondence between the sediment gradient and the velocity vector modulus in the velocity-deposition associated area is counted. According to the numerical correspondence, a constraint condition set including the coordinates of the associated area and the velocity and sediment matching rules is generated.

3. The method according to claim 2, characterized in that The second-order derivative distribution of the sediment gradient is superimposed on the primary velocity mutation area to screen out areas that simultaneously meet the extreme value judgment rule and where the sediment gradient has a sudden change as velocity-deposition associated areas, including: Based on the distribution of velocity vector lengths in the spatial units of the primary velocity mutation zone, a historical benchmark and a current cycle benchmark for the velocity vector length are set, and a primary screening area is defined in the spatial units that meet both the historical benchmark and the current cycle benchmark; Extracting sediment gradient data for each location point in the primary screening area, and superimposing and calculating the second-order derivative distribution of the sediment gradient in adjacent time periods for each location point; Setting parallel virtual measurement lines in the primary screening area along the water flow direction, generating a discrete point sequence at array intervals for each virtual measurement line, and mapping the second-order derivative distribution calculation results corresponding to the discrete point sequence into a second-order derivative intensity distribution; Based on the overall distribution state of the second-order derivative intensity distribution, an extreme value determination rule is set, and adjacent discrete point areas that continuously meet the extreme value determination rule are marked as sedimentation gradient mutation areas; The sediment gradient mutation zone is superimposed in three-dimensional space along the vertical direction of the virtual measurement line, and the spatial coordinates of the primary screening area and the distribution range of the sediment gradient mutation zone are integrated. The set of superimposed spatial units that simultaneously meet the velocity vector length benchmark condition and the extreme value judgment rule is defined as the velocity deposition association area.

4. The method according to claim 1, wherein Based on the revised spatiotemporal map of sediment deposition, the corresponding three-dimensional sediment velocity field vector distribution is calculated in real time through the scattering intensity difference and phase shift trajectory of the multi-band pulse signal, including: It simultaneously transmits a multi-band pulse signal containing a low-frequency penetration signal and a high-frequency tracking signal. After receiving the reflected signal, it separates and generates a sediment density distribution map penetrating to the bottom of the riverbed and a sediment movement trajectory map on the surface. Extracting the sediment accumulation density change rate of each depth layer in the vertical direction from the sediment density distribution map, and combining it with the sediment amount gradient at the corresponding position in the revised sediment deposition time-space map to generate a vertical density gradient correlation profile; Performing three-dimensional spatial decomposition on the surface sediment movement trajectory map, and establishing a surface particle movement direction angle distribution map based on the spatial position offset of the reflection signal feature points within the continuous transmission cycle; The vertical density gradient correlation profile is spatially matched with the surface particle movement direction angle distribution map, and the surface movement direction angle data of the corresponding position is superimposed in the vertical density gradient mutation area. The corresponding three-dimensional sediment velocity field vector distribution is generated through vector direction transfer.

5. The method according to claim 4, characterized in that The vertical density gradient correlation profile is spatially matched with the surface particle movement direction angle distribution map, and the surface movement direction angle data of the corresponding position is superimposed in the vertical density gradient mutation area. The corresponding three-dimensional sediment velocity field vector distribution is generated by vector direction transfer, including: Matching the spatial position of each depth layer data of the vertical density gradient correlation profile with the plane coordinates of the surface particle movement direction angle distribution map to screen out the vertical density gradient mutation area corresponding to the plane coordinates; Extracting surface motion direction angle data of the plane coordinates corresponding to the vertical density gradient mutation area, and calculating the horizontal motion vector of each plane coordinate point based on the surface motion direction angle data; Determine the gradient direction change amount of each position point in the vertical density gradient mutation area, and calculate the corresponding vertical direction correction value based on the gradient direction change amount; After superimposing the vertical direction correction value on the horizontal motion vector, the vector is transferred layer by layer along the vertical density gradient change direction, and the vector direction is corrected based on the offset angle between the gradient direction and the surface direction angle; The corrected motion vectors of all depth layers in the vertical density gradient mutation area are spatially superimposed to generate a three-dimensional sediment velocity field vector distribution covering the corresponding plane coordinates.

6. The method according to claim 1, characterized in that Synchronously acquiring multi-dimensional scanning data of the riverbed topography from an underwater robot and global coverage data from satellite remote sensing to generate a spatiotemporal map of sediment deposition including a sediment gradient, and periodically revising the spatiotemporal map of sediment deposition based on changes in the sediment gradient, including: Set the underwater robot's scanning path for the riverbed topography, synchronously obtain sediment accumulation thickness and multi-dimensional scanning data at each scanning point, and synchronously obtain full-area coverage data from satellite remote sensing during the robot's scanning intervals; The thickness variation difference of the sediment accumulation thickness at each scanning point is calculated according to the continuous scanning cycle, and the ratio of the thickness variation difference to the horizontal spacing between adjacent scanning points is used as the sediment gradient. The multi-dimensional scanning data and the global coverage data are combined to generate a spatiotemporal map of sediment deposition including the gradient direction and sediment gradient; When the direction of the deposition gradient at any scanning point continuously deviates from the initial direction by more than a set angle, or the fluctuation amplitude of the deposition gradient exceeds a set ratio, the scanning path encryption of the area where the point is located is triggered; The sediment accumulation thickness is reacquired in the encrypted scanning area, and the value and direction of the sediment gradient are corrected based on the new sediment accumulation thickness data, overwriting the data of the corresponding area in the original space-time map.

7. The method according to claim 1, characterized in that According to the set of constraints, combined with the real-time water level monitoring data and target water level range of the water conservancy hub reservoir area, a dynamic adjustment parameter sequence of the flood discharge gate opening is calculated, including: Obtain real-time water level monitoring data from the reservoir area of ​​the water conservancy hub, calculate the deviation between the current water level and the median of the target water level interval, and predict the intensity of water level change trends in future periods based on historical water level change data; Analyzing the flow velocity threshold and the sedimentation gradient threshold of each associated area in the constraint condition set, converting the ratio of the current measured flow velocity to the flow velocity threshold into a flow velocity adjustment parameter, and converting the ratio of the current sedimentation gradient to the sedimentation gradient threshold into a sedimentation gradient adjustment parameter; Performing a weighted fusion of the flow velocity adjustment parameter and the sedimentation gradient adjustment parameter to generate initial adjustment parameter values ​​for each associated area, and applying a directional correction to the initial adjustment parameter values ​​based on the water level deviation direction and trend strength; The corrected adjustment parameter values ​​of each area are arranged in time series to ensure that the parameter changes in adjacent time periods do not exceed the safety threshold of the gate mechanical action. The parameter differences between spatially adjacent areas are smoothed to generate a dynamic adjustment parameter sequence for the flood discharge gate opening.

8. A water level regulation system for a water conservancy project, characterized in that: include: An acquisition module, which simultaneously acquires multi-dimensional scanning data of the riverbed topography from underwater robots and global coverage data from satellite remote sensing for high-sediment-content waters in the reservoir area of ​​the water conservancy hub and the downstream river section of the spillway, to generate a spatiotemporal map of sediment deposition including a sediment gradient, and periodically modifies the spatiotemporal map of sediment deposition based on changes in the sediment gradient; A calculation module deploys a radar array in the bend transition section of the spillway outlet. Based on the modified spatiotemporal map of sediment deposition, the module calculates the corresponding three-dimensional sediment velocity field vector distribution in real time by using the scattering intensity differences and phase shift trajectories of multi-band pulse signals. a generation module that dynamically couples the sediment gradient with the three-dimensional sediment velocity field vector distribution, identifies the associated region of velocity mutation and sediment deposition within the bend transition section, and generates a constraint condition set including the coordinates of the associated region and a matching rule between velocity and sediment; The calculation module calculates the dynamic adjustment parameter sequence of the flood discharge gate opening according to the set of constraints and in combination with the real-time water level monitoring data and the target water level range of the water conservancy hub reservoir area.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a water level regulation method for a water conservancy hub as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, a method for regulating the water level of a water conservancy project according to any one of claims 1 to 7 is implemented.

Images