Basin joint dispatch hydrology and hydrodynamic coupling flood simulation method and system

Abstract

The application discloses a kind of river basin joint scheduling hydrology hydrodynamic coupling flood simulation method and system, the method comprises: collecting multi-source river basin data and carrying out standardization pretreatment to obtain pretreatment river basin data;Hydrology and hydrodynamics model based on two-way coupling mechanism are used to carry out real-time coupling flood simulation calculation to pretreatment river basin data, to obtain coupling flood simulation result;By pre-set engineering scheduling priority matrix, linkage coefficient dynamic calculation formula, conflict resolution strategy tree, extreme working condition scheduling step and global collaborative constraint, the coupling flood simulation result is processed to output initial scheduling scheme;Based on flood control safety target, water resources utilization target and engineering safety target, the initial scheduling scheme is iteratively optimized using improved NSGA-II algorithm to output optimal scheduling scheme set;The method can improve the real-time and accuracy of hydrology hydrodynamic coupling, realize the quantification and precision of multi-engineering scheduling.

Description

Technical Field

[0001] This invention relates to the field of water conservancy engineering technology, and more specifically, to a method and system for simulating floods by coupling hydrological and hydrodynamic processes in a basin joint scheduling. Background Technology

[0002] With the increasing frequency of global climate change and extreme weather events, the need for joint scheduling of river basin flood control projects and flood simulation is becoming increasingly urgent. Accurate flood simulation and efficient scheduling of flood control projects are key to ensuring river basin flood control safety and promoting the sustainable use of water resources.

[0003] Existing technologies in the area of ​​"flood control engineering scheduling and flood simulation using hydrological and hydrodynamic models" mainly fall into three categories: The first type of scheme is the single-project scheduling and isolated flood simulation scheme: This scheme designs scheduling rules only for a single flood control project (such as a reservoir or sluice gate), and the hydrological model and the hydrodynamic model operate independently. Typically, the rainfall-runoff results output by the hydrological model need to be manually input into the hydrodynamic model as boundary conditions, resulting in a lack of real-time interaction between the two and the absence of dynamic feedback correction capabilities for model parameters. This scheme cannot effectively cope with the complexity and real-time requirements of flood evolution.

[0004] The second type of scheme is a simple linkage and weakly coupled flood simulation scheme involving multiple projects. Although this scheme incorporates various types of flood control projects, its scheduling logic is usually "sequential execution" (e.g., pre-release of water from reservoirs before opening gates), lacking coordination constraints and conflict resolution mechanisms between projects, which can easily lead to conflicting scheduling instructions under complex operating conditions. In addition, the coupled simulation between hydrological and hydrodynamic models only transmits static parameters (e.g., updating coupled data at fixed intervals), lacking standardized data exchange interfaces and clear data format conversion rules. It cannot reflect the feedback of flood evolution on project scheduling in real time, resulting in a large deviation between the simulation results and the actual flood process, usually exceeding 15%.

[0005] The third type of scheme is the single-objective scheduling and general model flood simulation scheme: the scheduling optimization of this scheme usually only takes "flood control safety" as the single objective (such as minimizing the inundated area), failing to take into account multiple objectives such as "water resource utilization" and "engineering safety", which may lead to water waste or engineering overload risks. At the algorithm level, it adopts the standard NSGA-II algorithm for optimization, simply introducing a penalty function to handle constraints, lacking algorithmic improvements for water conservancy scheduling scenarios. In terms of computation and system, the simulation system relies on CPU serial computing, without parallel design for solving the core equations of the hydrodynamic model, resulting in long simulation times for complex watersheds (usually exceeding 2 hours) and low computational efficiency; at the same time, the system is often bound to Windows system and foreign GIS platform, which cannot meet the needs of localization and real-time scheduling in key fields.

[0006] In summary, existing technologies generally suffer from the following drawbacks: 1. Poor coordination in joint scheduling: The scheduling logic of various flood control projects (reservoirs-flood storage and detention areas-gates-pumping stations) is isolated, lacking unified scheduling decision priority determination and conflict coordination and resolution rules. This easily leads to operational conflicts such as "the reservoir needs to release floodwater but the downstream gate has reached its limit" and "the flood storage and detention area is full but the upstream is still diverting floodwater." Moreover, the linkage coefficients of the projects are mostly empirical values ​​without dynamic calculation logic.

[0007] 2. Delayed coupling between hydrological and hydrodynamic models: The coupling method is mostly static parameter transfer, lacking specific mapping formulas from water level to model parameters. No improvements have been made to the confluence link of hydrological models (such as the Xin'anjiang model) to support dynamic parameter correction. Furthermore, there is no standardized data exchange interface and clear data format conversion rules between models, resulting in a large deviation between simulation results and actual flood processes.

[0008] 3. Lack of multi-objective optimization: Focusing only on a single flood control objective fails to balance multiple objectives such as flood control safety, water resource utilization, and engineering safety, which can easily lead to water waste or engineering overload risks.

[0009] 4. Weak computing and adaptability: It relies on CPU serial computing, and the simulation of complex watersheds takes more than 2 hours, which is difficult to meet the real-time scheduling requirements; moreover, the system is bound to Windows system and foreign GIS platform, which cannot meet the needs of localization and cross-platform application in key fields.

[0010] Therefore, existing technologies have significant shortcomings in terms of real-time performance, accuracy, coordination, multi-objective optimization, and computational efficiency in the joint scheduling of watershed flood control projects and flood simulation. There is an urgent need to propose a new method and system that can solve the above problems. Summary of the Invention

[0011] One objective of this invention is to provide a new technical solution for a watershed joint scheduling hydrological and hydrodynamic coupled flood simulation method and system.

[0012] According to a first aspect of the present invention, a method for simulating floods by coupled hydrological and hydrodynamic scheduling in a watershed is provided, the method comprising: Step S1: Collect multi-source watershed data and perform standardized preprocessing on the multi-source watershed data to obtain preprocessed watershed data; Step S2: Use a hydrological and hydrodynamic model based on a two-way coupling mechanism to perform real-time coupled flood simulation calculations on the preprocessed watershed data to obtain coupled flood simulation results; Step S3: Process the coupled flood simulation results using a pre-set engineering scheduling priority matrix, dynamic calculation formula for linkage coefficients, conflict resolution decision tree, extreme working condition scheduling steps, and global collaborative constraints to output an initial scheduling scheme; Step S4: Based on the flood control safety objective, water resource utilization objective, and engineering safety objective, the initial scheduling scheme is iteratively optimized using the improved NSGA-II algorithm to output the optimal scheduling scheme set; Step S5: Output the optimal scheduling scheme set and the coupled flood simulation results to the interactive interface for visualization.

[0013] Optionally, in step S1, the multi-source watershed data includes: basic data, real-time data, and scheduling rule data; The basic data specifically includes: DEM topographic data, land use data, river cross-section data, and flood control engineering parameter data; the real-time data specifically includes: real-time data from rainfall stations, real-time data from water level stations, and real-time data from flow stations; the scheduling rule data specifically includes: reservoir flood control scheduling curve data, flood storage and detention area flood control scheduling plan data, gate graded control rule data, and pump station linkage drainage logic rule data.

[0014] Optionally, in step S2, the hydrological and hydrodynamic model includes a hydrological sub-model and a hydrodynamic sub-model; Step S2 specifically includes: Step S21: The hydrological sub-model uses the improved Xin'anjiang model to calculate the runoff generation and confluence process of the preprocessed watershed data to obtain the hydrological model calculation results that include the river inflow process; Step S22: The hydrodynamic sub-model is based on the two-dimensional Saint-Venant equations. The finite volume method is used to solve the hydrodynamic model calculation results and the topographic data contained in the preprocessed watershed data to obtain the hydrodynamic model calculation results containing real-time water level and velocity distribution. Step S23: Select key cross sections within the watershed as feedback nodes. Extract the real-time water level value and predicted water level value of the key cross section from the calculation results of the hydrodynamic model and the calculation results of the hydrological model, respectively. Calculate the cross section water level deviation between the real-time water level value and the predicted water level value of the key cross section. When the cross section water level deviation exceeds a preset threshold, correct the roughness parameter of the confluence link in the hydrological sub-model according to the preset mapping formula between water level deviation and roughness, so as to realize the dynamic update of the parameters of the hydrological sub-model.

[0015] Optionally, in step S21, the improved Xin'anjiang model specifically involves: using the improved Muskingum method for flow calculation to achieve the correlation mapping between the roughness value and the Muskingum method parameters; The specific formula for the improved Muskingan method is expressed as follows: (1) in, The improved outflow rate at the end of the time period after roughness correction; This represents the inflow rate at the end of the time period; This represents the inflow rate at the beginning of the time period; The outflow rate at the beginning of the time period; , , The roughness-corrected Muskingen calculus coefficients are expressed as follows: In the formula For time step; These are the Muskingan method storage coefficient and flow rate specificity factor after roughness correction, respectively, which are linearly related to the roughness value n, and the correlation formula is: , , The initial storage capacity coefficient and initial flow rate weight factor are for the Muskingan method. The initial value of roughness, This is the corrected roughness value.

[0016] Optionally, in step S22, a multi-level parallel strategy of river segmentation and two-dimensional grid partitioning is adopted to distribute the solution calculation task to the multi-stream processor of the GPU for processing, so as to achieve computational acceleration.

[0017] Optionally, in step S23, the roughness parameter of the confluence link in the hydrological sub-model is corrected according to a preset mapping formula between water level deviation and roughness, specifically as follows: The cross-sectional water level deviation is quantitatively converted into a roughness correction coefficient using a piecewise linear preset mapping formula. Then, the product of the roughness correction coefficient and the initial roughness value is calculated to obtain the roughness correction value. Based on the principle of nearest correction in sub-basins, the roughness parameters of the confluence link in the hydrological sub-model are corrected according to the roughness correction value, so as to realize the dynamic updating of the parameters of the hydrological sub-model. The piecewise linear preset mapping formula is expressed as follows: (2) In the formula, This is the roughness correction coefficient, derived from the cross-sectional water level deviation. The only certainty is that m is the unit of length, the meter.

[0018] Optionally, step S4 specifically includes: Step S41: Initialize the population of scheduling parameters using the Latin hypercube sampling method. The scheduling parameters include the reservoir pre-discharge coefficient, flood diversion priority of flood storage and detention areas, gate opening adjustment coefficient, and pumping station drainage coefficient. Step S42: Construct a three-objective optimization function based on flood control safety objective, water resource utilization objective, and engineering safety objective, and calculate the total objective function value of each individual in the population by combining the water conservancy scheduling-specific nonlinear penalty function, wherein the water conservancy scheduling-specific nonlinear penalty function is used to quantify engineering constraints into penalty values; Step S43: Perform a fast non-dominated sort on the population, and calculate the weighted crowding degree for each individual using a weighted crowding degree calculation method, wherein the weighted crowding degree assigns a higher weight to the flood control safety target than to the water resource utilization target and the engineering safety target; Step S44: Perform genetic evolution operations on the population using adaptive crossover mutation probability, wherein the adaptive crossover mutation probability is dynamically adjusted according to the overall fitness of individuals in the population; Step S45: Perform a hill-climbing algorithm local search operation on each individual after the genetic evolution operation to search for a better solution in the parameter neighborhood; Step S46: Determine whether the iteration termination condition is met. If it is met, output the optimal scheduling optimal solution set. If it is not met, return to step S42 and repeat until the termination condition is met.

[0019] According to a second aspect of the present invention, a watershed joint scheduling hydro-hydraulic coupling flood simulation system is provided, the system comprising: The data module is configured to collect multi-source watershed data and perform standardized preprocessing on the multi-source watershed data to obtain preprocessed watershed data. The coupled calculation module is configured to use a hydrological and hydrodynamic model based on a two-way coupling mechanism to perform real-time coupled flood simulation calculations on the preprocessed watershed data to obtain coupled flood simulation results. The joint scheduling module is configured to process the coupled flood simulation results through a pre-set engineering scheduling priority matrix, a dynamic calculation formula for linkage coefficients, a conflict resolution decision tree, extreme working condition scheduling steps, and global collaborative constraints, so as to output an initial scheduling scheme. The multi-objective optimization module is configured to iteratively optimize the initial scheduling scheme based on flood control safety objectives, water resource utilization objectives, and engineering safety objectives using an improved NSGA-II algorithm, so as to output an optimal scheduling scheme set. The interactive module is configured to output the optimal scheduling scheme set and the coupled flood simulation results to the interactive interface for visualization.

[0020] According to a third aspect of the present invention, an electronic device is provided, the electronic device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the watershed joint scheduling hydrological and hydrodynamic coupled flood simulation method as described in the first aspect of the present invention.

[0021] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the watershed joint scheduling hydrological-hydraulic coupled flood simulation method described in the first aspect of the present invention.

[0022] According to an embodiment disclosed in this invention, the watershed joint scheduling hydrological-hydraulic coupled flood simulation method and system of this invention has the following beneficial effects: This invention discloses a watershed joint scheduling hydrological-hydrodynamic coupled flood simulation method and system. It employs a conflict resolution and quantitative scheduling scheme for multi-project joint scheduling, including a multi-project scheduling priority matrix, a conflict resolution decision tree, step-by-step scheduling logic for extreme conditions (reservoir discharge but flood storage area is full), a dynamic calculation formula for project linkage coefficients, and a quantitative adjustment formula for global collaborative constraints, achieving precise coordination of multi-project scheduling. Furthermore, it utilizes a real-time hydrological-hydrodynamic coupling strategy, including a piecewise linear preset mapping formula for water level deviation to roughness correction, a parameter correlation formula for the improved Muskingan method in the Xin'anjiang model, and standardized data exchange from the hydrological-hydrodynamic model. The interface definition was changed to address the issue of the original model's inability to dynamically adjust parameters. A dedicated, improved NSGA-II algorithm was adopted for water conservancy scheduling, featuring five substantial improvements that distinguish it from the standard NSGA-II algorithm: Latin hypercube sampling for population initialization, a formula for constructing a dedicated nonlinear penalty function for water conservancy scheduling, an adaptive crossover and mutation probability calculation method, a weighted congestion calculation formula, and a hill-climbing algorithm for local search after genetic evolution. This enhances the engineering applicability of the optimization results. A GPU parallel acceleration scheme for solving the Saint-Venant equations was adopted, employing a multi-level parallel strategy of "river channel segmentation + two-dimensional grid partitioning," a thread block-thread mapping method, and a memory optimization strategy of batch transmission / memory reuse / data block caching to achieve efficient acceleration of hydrodynamic calculations. A bidirectional driving architecture for engineering scheduling and coupled simulation was implemented, with a real-time data interaction design between the joint scheduling layer and the coupled calculation layer. Dynamic adjustments of the coupled calculation parameters by scheduling commands under extreme conditions ensure that the scheduling scheme is dynamically adjusted based on the latest flood evolution results.

[0023] Other features and advantages of the invention will become clear from the following detailed description of exemplary embodiments of the invention with reference to the accompanying drawings. Attached Figure Description

[0024] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments of the invention and, together with their description, serve to explain the principles of the invention.

[0025] Figure 1 This is a schematic flowchart of a watershed joint scheduling hydro-hydrodynamic coupled flood simulation method according to an embodiment; Figure 2 This is another schematic diagram of a watershed joint scheduling hydro-hydrodynamic coupled flood simulation method provided in the embodiments; Figure 3 This is a logic diagram for the joint scheduling of multiple types of flood control projects according to the embodiments; Figure 4 This is a schematic diagram illustrating the specific process of the multi-objective optimization algorithm provided in the embodiment; Figure 5 This is a structural block diagram of a watershed joint scheduling hydro-hydrodynamic coupled flood simulation system provided according to an embodiment; Figure 6 This is an architecture diagram of a watershed joint scheduling hydro-hydrodynamic coupled flood simulation system provided according to an embodiment; Figure 7 This is a schematic diagram of an electronic device. Detailed Implementation

[0026] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention.

[0027] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use.

[0028] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0029] In all the examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. Example 1:

[0030] See Figure 1 As shown in the figure, this embodiment of the invention provides a watershed joint scheduling hydrological and hydrodynamic coupled flood simulation method, which includes: Step S1: Collect multi-source watershed data and perform standardized preprocessing on the multi-source watershed data to obtain preprocessed watershed data.

[0031] Optionally, in step S1 of the watershed joint scheduling hydrological-hydrodynamic coupled flood simulation method of this embodiment, the multi-source watershed data includes: basic data, real-time data, and scheduling rule data; wherein, the basic data specifically includes: DEM topographic data, land use data, river cross-section data, and flood control engineering parameter data; the real-time data specifically includes: real-time data from rainfall stations, real-time data from water level stations, and real-time data from flow stations; the scheduling rule data specifically includes: reservoir flood control scheduling curve data, flood storage and detention area flood control scheduling plan data, gate graded control rule data, and pump station linkage drainage logic rule data.

[0032] For details, see Figure 2 As shown, the first step in the watershed joint scheduling hydrological and hydrodynamic coupled flood simulation method of this embodiment is to collect multi-source watershed data and perform standardized preprocessing. The preprocessed data, i.e., the preprocessed watershed data, are all converted into the water conservancy industry standard SDTS format (Spatial Data Transfer Standard), providing a unified basis for subsequent data exchange between models.

[0033] (1) Data collection scope Basic data: DEM (Digital Elevation Model) topographic data (resolution 10m~100m), land use data, river cross-section data, flood control engineering parameters (total reservoir capacity / flood control capacity, flood storage and detention area activation threshold / flood diversion capacity, gate size / maximum opening, pumping station drainage flow / start and stop threshold).

[0034] Real-time data: Rainfall station data (e.g., intervals of 5-15 minutes), water level station data (e.g., intervals of 10 minutes), and flow rate station data (e.g., intervals of 30 minutes) are accessed via the MQTT (Message Queuing Telemetry Transport) protocol.

[0035] Dispatch rule data: reservoir flood control dispatch curves (e.g., flood season limit water level, flood control high water level), flood diversion dispatch plans for flood storage and detention areas, gate graded control rules, and pump station linkage drainage logic, stored in JSON (JavaScript Object Notation) standard format.

[0036] (2) Preprocessing process The GDAL library (Geospatial Data Abstraction Library) converts data in different formats (e.g., Shapefile, SHP; Tagged Image File Format, TIFF; Comma-Separated Values, CSV) into a unified GeoJSON format. The PROJ library is used to unify the coordinate system to CGCS2000; the Spdlog log library is used to record data quality (outlier removal adopts the 3σ principle, with a removal rate of ≤0.3%); finally, the preprocessed standardized data is stored in the SQLite3 database to form a structured dataset of "data identifier-collection time-data value-quality level".

[0037] Step S2: Use a hydrological and hydrodynamic model based on a two-way coupling mechanism to perform real-time coupled flood simulation calculations on the preprocessed watershed data to obtain coupled flood simulation results.

[0038] Optionally, in step S2 of the watershed joint scheduling hydro-hydrodynamic coupled flood simulation method of this embodiment, the hydrological and hydrodynamic model includes a hydrological sub-model and a hydrodynamic sub-model; step S2 specifically includes: Step S21: The hydrological sub-model uses the improved Xin'anjiang model to calculate the runoff generation and confluence process of the preprocessed watershed data to obtain the hydrological model calculation results that include the river inflow process; Step S22: The hydrodynamic sub-model is based on the two-dimensional Saint-Venant equations. The finite volume method is used to solve the hydrodynamic model calculation results and the topographic data contained in the preprocessed watershed data to obtain the hydrodynamic model calculation results containing real-time water level and velocity distribution. Step S23: Select key sections within the watershed as feedback nodes. Extract the real-time water level and predicted water level of the key sections from the hydrodynamic model calculation results and the hydrological model calculation results, respectively. Calculate the cross-sectional water level deviation between the real-time water level and the predicted water level. When the cross-sectional water level deviation exceeds a preset threshold, correct the roughness parameter of the confluence link in the hydrological sub-model according to the preset mapping formula between water level deviation and roughness, so as to achieve dynamic updating of the parameters of the hydrological sub-model.

[0039] Optionally, in step S21 of the watershed joint scheduling hydro-hydrodynamic coupled flood simulation method in this embodiment, the improved Xin'anjiang model specifically uses the improved Muskingum method for confluence calculation to achieve the correlation mapping between roughness value and Muskingum method parameters; The specific formula for the improved Muskingan method is expressed as follows: (1) in, The improved outflow rate at the end of the time period after roughness correction; Inflow at the end of the period quantity; This represents the inflow rate at the beginning of the time period; The outflow rate at the beginning of the time period; , , The roughness-corrected Muskingen calculus coefficients are expressed as follows: In the formula For time step; These are the Muskingan method storage coefficient and flow rate specificity factor after roughness correction, respectively, which are linearly related to the roughness value n, and the correlation formula is: , , The initial storage capacity coefficient and initial flow rate weight factor are for the Muskingan method. The initial value of roughness, This is the corrected roughness value.

[0040] Optionally, in step S22, the watershed joint scheduling hydrological and hydrodynamic coupled flood simulation method of this embodiment adopts a multi-level parallel strategy of river segmentation and two-dimensional grid segmentation to allocate the solution calculation task to the multi-stream processor of the GPU for processing, so as to achieve computational acceleration.

[0041] Optionally, in step S23 of the watershed joint scheduling hydro-hydrodynamic coupled flood simulation method of this embodiment, the roughness parameter of the confluence link in the hydrological sub-model is corrected according to the preset mapping formula between water level deviation and roughness, specifically as follows: The cross-sectional water level deviation is quantitatively converted into a roughness correction coefficient using a piecewise linear preset mapping formula. Then, the product of the roughness correction coefficient and the initial roughness value is calculated to obtain the roughness correction value. Based on the principle of nearest correction in sub-basins, the roughness parameters of the confluence link in the hydrological sub-model are corrected according to the roughness correction value, so as to realize the dynamic updating of the parameters of the hydrological sub-model. The piecewise linear pre-defined mapping formula is expressed as follows: (2) In the formula, This is the roughness correction coefficient, derived from the cross-sectional water level deviation. The only certainty is that m is the unit of length, the meter.

[0042] For details, see Figure 2 As shown in this embodiment, the real-time coupled flood simulation calculation process of the hydrological and hydrodynamic models in the watershed joint scheduling hydro-hydrodynamic coupled flood simulation method is as follows: First, a two-way coupling mechanism of "hydrological runoff generation and hydrodynamic evolution" is constructed, with a coupling cycle of 30 minutes per cycle, synchronizing the calculation step size of the hydrological and hydrodynamic models (i.e., the corresponding...). Figure 2 (Closed loop of "hydrological calculation - hydrodynamic calculation - feedback correction") (1) Hydrological sub-model calculation: The improved Xin'anjiang model is adopted. Based on the original runoff generation and confluence calculation, the Muskingan method of the confluence link is improved to realize the correlation mapping between the roughness coefficient n value and the Muskingan method parameters K (initial storage coefficient, i.e., confluence time) and X (discharge weight coefficient), which solves the problem that the original Xin'anjiang model cannot dynamically adjust the confluence parameters. Input the standardized data (time period net rainfall, watershed underlying surface parameters) after the preprocessing in the previous step, calculate the runoff generation and confluence process, output the river inflow process (1-hour step), and obtain the hydrological model calculation results.

[0043] Production flow calculation: (3) in, The output flow rate (in mm) Rainfall amount (in mm), Initial loss (unit: mm) The stable permeability rate (unit: mm / h) The duration of the flow (in hours).

[0044] The flow calculation employs a modified Muskingan method, introducing a dynamic correction for the roughness coefficient n: Revised formula: (1) in, The improved outflow rate at the end of the time period after roughness correction; Inflow at the end of the period quantity; This represents the inflow rate at the beginning of the time period; The outflow rate at the beginning of the time period; , , The roughness-corrected Muskingen calculus coefficients are expressed as follows: In the formula For time step; These are the Muskingan method storage coefficient and flow rate specificity factor after roughness correction, respectively, which are linearly related to the roughness value n, and the correlation formula is: , , The initial storage capacity coefficient and initial flow rate weight factor are for the Muskingan method. The initial value of roughness, This is the corrected roughness value.

[0045] (2) Hydrodynamic sub-model calculation: Based on the two-dimensional Saint-Venant equations, the inflow process of the hydrological model and DEM topographic data are input, and the finite volume method is used to solve the problem. The water level and velocity distribution over a time period (30-minute step) are calculated. The solution process is accelerated by GPU parallel processing. The solution formula is as follows: h / t + (hu) / x + (hv) / y = q (4) Where h is the water depth, u and v are the flow velocities in the x and y directions, and q is the source and sink term; (hu) / t + (hu² + gh² / 2) / x + (huv) / y = -gh z / x + τ_bx / ρ + f_v(5) Where z is the riverbed elevation, τ_bx is the shear stress on the riverbed, ρ is the water density, and f_v is the viscous force.

[0046] (3) Real-time coupled feedback By using a specific mathematical mapping and correction logic from water level to roughness, quantitative feedback correction of the water level calculated by the hydrodynamic sub-model on the confluence parameters of the hydrological sub-model is achieved.

[0047] Step 1: Calculation of water level deviation at key sections: Select the control sections where hydrological stations / water level stations are located within the basin as feedback nodes, and calculate the deviation between the values ​​calculated by the hydrodynamic sub-model and the predicted values ​​by the hydrological sub-model. (6) in, The water level (in meters) at the key cross-section calculated for the hydrodynamic sub-model. Set a deviation threshold for the water level (in meters) at the key cross-section predicted by the hydrological sub-model. The correction will be initiated at that time.

[0048] Step 2: Specific mapping formula from water level deviation to roughness correction (f(H)→n): A piecewise linear preset mapping formula is used to achieve the quantitative conversion from water level deviation to roughness correction coefficient, replacing the qualitative division of the original fuzzy adaptive algorithm, resulting in a more accurate mathematical expression. (2) Among them, the roughness correction value is , This is the roughness correction coefficient, derived from the water level deviation. The only certainty, The initial value for roughness (assigned according to land use type). This is the corrected roughness value.

[0049] Step 3: Spatial range of parameter correction: In this embodiment, the parameters of the entire watershed are not corrected. Instead, the principle of correction in the nearest sub-watershed is adopted: taking the key section that triggers the correction as the center, the sub-watershed where it is located and 1-2 adjacent sub-watersheds upstream and downstream are selected as the correction range. The roughness initial value of the remaining sub-watersheds remains unchanged to avoid simulation distortion caused by correction of the entire watershed.

[0050] Step 4: Feedback of corrected parameters: Feedback of corrected roughness parameters Substituting the correlation formula (1) of the improved Muskingan method above, we can calculate the result. The data is then fed back to the confluence calculation stage of the Xin'anjiang model to update the model parameters and complete one coupling feedback.

[0051] The coupling cycle can be set to 30 minutes per cycle, which is the calculation step size of the synchronous hydrological and hydrodynamic models.

[0052] (4) Definition of data exchange interface between models A standardized data exchange interface for hydrological and hydrodynamic models was established, clarifying the format, content, and frequency of data transmission to address the lack of rules in data exchange in the original technology. The interface definition is shown in Table 1 below: Table 1: Data format conversion rules: When converting the vector data (inflow process) of the hydrological sub-model to the grid surface data of the hydrodynamic sub-model, the Thiessen polygon interpolation method is used; when converting the grid data (water level) of the hydrodynamic sub-model to the point data of the hydrological sub-model, the grid centroid sampling method is used.

[0053] (5) Computing power support By utilizing GPU parallel computing (CUDA architecture), a highly efficient parallel acceleration of the solution to the Saint-Venant equations is achieved, resulting in a computational efficiency improvement of 4-6 times compared to CPU serial computation. Specific implementation details of GPU acceleration are as follows: Parallel Strategy: A multi-level parallel strategy of segmenting the river channel and dividing it into two-dimensional grid blocks is adopted. For the grid data of the two-dimensional hydrodynamic calculation of the watershed, parallel mapping is performed based on the CUDA Thread Block-Thread two-level thread model, and the computing tasks are distributed to the multi-stream processors of the GPU.

[0054] Mesh generation and thread mapping: The overall two-dimensional computational grid of the watershed is divided into several sub-grid blocks of size 32×32 (matching the optimal size of CUDA Thread Block), and each sub-grid block is mapped to a GPU ThreadBlock; the grid cells within each sub-grid block are mapped to a GPU Thread, and each Thread is responsible for solving the Saint-Venant equations for one grid cell; For river-type watersheds, the watersheds are further segmented according to the direction of the river, and each segment is assigned to an independent GPU stream to achieve asynchronous parallel computing between the watersheds.

[0055] Memory optimization strategy (reducing CPU-GPU data transfer latency): Batch transfer: Transfer all input data (DEM, initial water level, inflow process) within a 30-minute computation cycle from the CPU host memory to the GPU global memory at once, avoiding frequent small batch transfers. Memory reuse: Utilize intermediate variables (such as flow rate) from solving the Saint-Venant equations. , water depth The data is stored in GPU shared memory for shared access by threads within a Thread Block, replacing frequent reads and writes to global memory. Data block caching: GPU memory is divided into input buffers, computation buffers, and output buffers to store input data, intermediate results, and output results respectively. The buffer size is dynamically adjusted according to the sub-grid block size to avoid memory waste. Parallel solution process: After CPU preprocessing, standardized data is transferred in batches to GPU memory → GPU solves the equation system of sub-grid blocks in parallel according to Thread Block-Thread mapping → The solution results are temporarily stored in GPU shared memory. After completing one computation step, key results (water level, flow velocity) are returned to CPU host memory → used for parameter correction and engineering scheduling calculations in the hydrological model.

[0056] Step S3: Process the coupled flood simulation results using a pre-defined engineering scheduling priority matrix, dynamic calculation formula for linkage coefficients, conflict resolution decision tree, extreme condition scheduling steps, and global collaborative constraints to output an initial scheduling scheme.

[0057] For details, see Figure 3 As shown in this embodiment, the joint scheduling execution process of multiple flood control projects in the watershed joint scheduling hydro-hydrodynamic coupled flood simulation method is as follows: Based on the above coupled flood simulation results, the coordinated scheduling and conflict resolution of multiple projects are achieved through the engineering scheduling priority matrix, the dynamic calculation formula of the linkage coefficient, the conflict resolution decision tree, and the specific scheduling steps for extreme conditions, as follows: (1) Multi-project scheduling priority matrix Priority rules for the scheduling of flood control projects in the basin were formulated. Following the principle of "flood control safety first, project safety second, and water resource utilization last," a four-level project scheduling priority matrix was constructed to clarify the scheduling weights of different projects, providing a basis for conflict resolution, as shown in Table 2. Table 2: Priority rule: When scheduling instructions for projects of different priorities conflict, the scheduling instructions for the higher priority project are executed first, and the lower priority project is dynamically adjusted according to the execution result of the higher priority project; when scheduling instructions for projects of the same priority conflict, the conflict resolution decision tree is used to determine the outcome.

[0058] (2) Scheduling logic for each project Reservoir scheduling: When the inflow is greater than or equal to the flood control standard flow (e.g., once in 20 years), pre-discharge is carried out according to the "flood control scheduling curve". The pre-discharge flow = (inflow - downstream safe discharge) × 0.8 (reserve safety factor); if the downstream river level exceeds the warning level, the pre-discharge flow will be automatically reduced to prioritize downstream safety. Triggering condition: Inbound traffic Q in ≥ flood control standard flow rate Q std (e.g., a 20-year flood event), or the downstream river level H down ≥Warning water level H warn ; Scheduling logic: When only Q in ≥Q std At that time, the pre-discharge flow rate Q release = (Q in - Q safe )×0.8(Q safe (where H is the downstream safe discharge capacity, and 0.8 is the safety factor); down ≥H warn At that time, Q release = Q release ×0.6 (Dynamically reduce leakage); Coordination constraint: Pre-discharge flow rate ≤ maximum discharge capacity of the reservoir Q max And the outflow rate is less than or equal to the downstream river channel flow capacity Q. river .

[0059] Flood storage and detention area operation: When the water level at the river control section is greater than or equal to the flood diversion threshold, the flood storage and detention area is activated according to the "flood diversion priority" (e.g., low-lying areas are given priority). The flood diversion volume = (actual water level at the section - flood diversion threshold water level) × river flow coefficient. At the same time, the pumping station linkage is triggered (the pumping station starts to drain water within 30 minutes after the flood diversion, and the drainage flow rate = flood diversion volume × 0.6 to avoid water accumulation in the flood storage and detention area). Triggering condition: Water level H at the control section ctrl ≥ flood diversion threshold H divAnd after the reservoir pre-releases water, H ctrl Still ≥H di ; Scheduling logic: Sort by the degree of low elevation (priority P1) P2 P3), flood diversion volume Q div = (H ctrl - H div )×K river (K) river (This is the river flow coefficient, ranging from 0.8 to 1.2, dynamically adjusted according to the river roughness). Coordination constraint: Flood diversion volume ≤ Total capacity of flood storage and detention area V div The activation order complies with the flood control plan; the pump station linkage is triggered within 30 minutes after the flood diversion.

[0060] Gate operation: The gate opening is dynamically adjusted according to the water level in different sections of the river. When the water level is at the warning level, the opening should be maintained at 50% (while also allowing navigation); when the water level is ≥ the warning level and When the water level is guaranteed, the gate opening is adjusted to 100% for flood discharge; when the water level is ≥ the guaranteed water level, the upstream reservoir is coordinated to reduce the discharge, and the backup gate is opened at the same time. Triggering condition: River water level H sec H warn At that time, the opening degree α = 50%; H war ≤H sec H guar (When the water level is guaranteed), α = 100%; H sec ≥H guar At that time, emergency coordination was initiated with upstream projects; Scheduling logic: The opening degree is adjusted in a step-by-step manner, with a single adjustment range of ≤20%, to avoid sudden changes in water level; Cooperative constraints: The maximum gate opening α_max ≤ 1.2 m, and the rate of change of river level after the opening adjustment ≤ 0.1 m / h.

[0061] Pump station dispatching: Triggering conditions: Within 30 minutes after the flood diversion of the flood storage and detention area, and the water level H in the flood storage and detention area... det ≥H pump ; Dispatch logic: Improve the linkage coefficient of the empirical value, changing the empirical coefficient 0.6 of "pump station drainage flow = flood diversion volume × 0.6" to a dynamic calculation coefficient K. p The linkage coefficient is determined by both the water level difference in the flood storage and detention area and the installed capacity of the pumping station, enabling quantitative calculation of the linkage coefficient. Pump station drainage flow rate H pump Dynamic formula: (7) Among them, the dynamic coefficient is: K p H is the dynamic coefficient for drainage at the pumping station (range 0.4~0.8). det H represents the actual water level (m) in the flood storage and detention area. pump H is the drainage threshold (m) for the pumping station. div P is the flood threshold (m) for flood storage and detention areas. act P represents the actual installed capacity (kW) of the pumping station. max This refers to the total installed capacity of the pumping station (kW).

[0062] Collaborative constraints: The continuous operating time of the pumping station is ≤72 hours, and Q pump ≤ Total installed capacity of pumping station P max .

[0063] (3) Decision tree for resolving conflicts in multi-project scheduling To address scheduling conflict issues among projects of the same priority, a decision tree for resolving flood control project scheduling conflicts is constructed. Employing a logic of "layered judgment, quantitative calculation, and step-by-step adjustment," the scheduling priority in case of conflict is clearly defined. The core judgment steps of the decision tree are as follows: The first level of assessment is flood risk level assessment: Calculate the basin flood risk level R (R=H) calc / H guar H guar To ensure water level stability, a water level of R ≥ 1.0 indicates high risk, and the "flood discharge / diversion" command should be prioritized; a water level of 0.8 ≤ R 1.0 is classified as medium risk, balancing flood discharge and engineering safety; R A value of 0.8 indicates low risk and allows for consideration of water resource utilization.

[0064] The second layer of judgment is the engineering safety margin judgment: calculate the upper limit of the safety margin parameter of the conflicting engineering and the actual parameter upper limit S = (parameter upper limit – actual parameter) / parameter upper limit. Prioritize the execution of the scheduling instructions of the engineering with higher safety margin to avoid engineering overload.

[0065] The third layer determines the impact range of the flood peak: calculate the impact range A of engineering scheduling on the flood peak, and prioritize the execution of scheduling instructions for engineering projects with a larger impact range to maximize flood control benefits.

[0066] (4) Specific scheduling steps for extreme operating conditions For the extreme situation where the reservoir must release floodwater but the downstream flood storage area is already full, a detailed calculation and scheduling logic in 8 steps is given to ensure the feasibility of engineering scheduling. This situation is a high-risk situation in the basin (R≥1.0). The first priority project is the pumping station, and the second priority project is the reservoir / main gate. Operating condition premise: Reservoir inflow rate Q inIf the reservoir's water level exceeds twice the standard flood control flow rate and the flood control high water level is exceeded, flood discharge is necessary; the actual water level H in the downstream flood storage and detention area... det = H div-max (Water level corresponding to the maximum capacity of the flood storage and detention area), with no flood diversion space.

[0067] The first step is data collection and risk assessment: collecting real-time water levels / discharge capacity of reservoirs, gate openings / safety margins of downstream main river channels, pumping station operating capacity / drainage capacity, and water levels at key river sections, and calculating the basin flood risk level R=H. calc / H guar If the value is ≥1.0, it is determined to be a high-risk operating condition, and the extreme operating condition scheduling process is initiated.

[0068] The second step is to calculate the engineering safety margin: calculate the safety margin S of the downstream main channel gates. gate Safety margin S of the pumping station pump The formula is S = (parameter upper limit – actual parameter) / parameter upper limit.

[0069] The third step is to issue a full-load drainage command to the pumping station: This command is issued to the pumping station according to the highest priority level. ( To maximize the efficiency of the pumping station (take 0.85), we aim to reduce the water level in the flood storage area and downstream river channels.

[0070] Step 4: Optimize the opening of the gates in the outer main channel: Calculate the maximum safe opening of the gates in the downstream main channel according to the second priority level. The formula is Adjust the gate opening to This will improve the river's flow capacity.

[0071] The fifth step is to quantitatively calculate the reservoir's flood discharge flow: Based on the maximum flow capacity of the downstream river channel, calculate the reservoir's maximum safe flood discharge flow. ( (This is the maximum flow capacity of the downstream main channel), ensuring that the total flow of the downstream channel does not exceed its maximum flow capacity after the reservoir discharges floodwater.

[0072] The sixth step is to implement tiered flood discharge from the reservoir: the reservoir discharge flow is divided into three levels, according to "0.3×Q". release-max →0.6×Q release-max →Q release-max The flood discharge volume is increased in stages in sequence, with each stage spaced 10 minutes apart, to avoid a sudden increase in the flood peak.

[0073] The seventh step is real-time monitoring and dynamic adjustment: collect the water level of key sections of the downstream river every 5 minutes. If the water level is declining, maintain the current flood discharge / dispatch instructions; if the water level is still rising, activate the tributary gates to assist in flood discharge and open all gates of the tributary river to 100%.

[0074] Step 8 is the condition termination determination: when the reservoir water level drops below the flood control high water level, and the water level at the key section of the downstream river channel is H... calc H guar When the extreme working condition scheduling process is terminated, normal collaborative scheduling is restored.

[0075] (5) Global Coordination Constraints: Set "engineering safety thresholds" (e.g., maximum gate opening ≤ 1.2m, continuous pump station operation time ≤ 72 hours). When a certain project approaches the threshold, the adjustment range and calculation logic for other projects are as follows: When the safety margin S of any project is less than or equal to 0.1, the adjustment range ∆% of other projects is calculated using the following formula: ∆% = 20% × (0.1 - S) / 0.1, where ∆% is the adjustment range (ranging from 0 to 20%) and S is the safety margin of projects approaching the threshold.

[0076] Monitoring mechanism: Real-time collection of various project operation parameters (reservoir discharge, flood diversion volume in flood storage and detention areas, gate opening, pumping station flow) and calculation of safety margin S = (parameter upper limit - actual parameter) / parameter upper limit; Linkage logic: When any project S≤0.1 (close to the safety threshold), other projects are automatically adjusted: if the gate S=0.08 (close to the maximum opening), then ∆%=4%, the flood diversion volume in the flood storage and detention area increases by 4%, and the pre-discharge volume of the reservoir decreases by 4%; if the pumping station S=0.05 (close to the upper limit of the operating time), then ∆%=10%, the flood diversion volume in the flood storage and detention area decreases by 10%, and the gate opening increases by 10%.

[0077] Step S4: Based on the flood control safety objectives, water resource utilization objectives, and engineering safety objectives, the initial scheduling scheme is iteratively optimized using the improved NSGA-II algorithm to output the optimal scheduling scheme set.

[0078] Optionally, step S4 in the watershed joint scheduling hydro-hydrodynamic coupled flood simulation method of this embodiment of the invention specifically includes: Step S41: Initialize the population of scheduling parameters using the Latin hypercube sampling method. The scheduling parameters include the reservoir pre-discharge coefficient, flood diversion priority of flood storage and detention areas, gate opening adjustment coefficient, and pumping station drainage coefficient. Step S42: Construct a three-objective optimization function based on flood control safety objectives, water resource utilization objectives, and engineering safety objectives, and calculate the total objective function value of each individual in the population by combining the water conservancy scheduling-specific nonlinear penalty function. The water conservancy scheduling-specific nonlinear penalty function is used to quantify engineering constraints into penalty values. Step S43: Perform a fast non-dominated sort on the population and calculate the weighted crowding degree for each individual using a weighted crowding degree calculation method. The weighted crowding degree assigns a higher weight to the flood control safety target than to the water resource utilization target and the engineering safety target. Step S44: Perform genetic evolution operations on the population using adaptive crossover mutation probability, wherein the adaptive crossover mutation probability is dynamically adjusted according to the overall fitness of individuals in the population; Step S45: Perform a hill-climbing algorithm local search operation on each individual after the genetic evolution operation to search for a better solution in the parameter neighborhood; Step S46: Determine whether the iteration termination condition is met. If it is met, output the optimal scheduling optimal solution set; if not, return to step S42 and repeat until the termination condition is met.

[0079] Specifically, in this embodiment, the watershed joint scheduling hydro-hydrodynamic coupled flood simulation method is based on... Figure 4 The process of "population initialization - genetic operation - local search - non-dominated sorting" is followed by an improved NSGA-II algorithm (introducing an engineering constraint penalty factor) to iteratively optimize the initial scheduling scheme.

[0080] This embodiment focuses on the scenario of water conservancy flood control scheduling. The NSGA-II algorithm has made five substantial improvements, as shown in Table 3: Table 3: ② Water resource utilization objective (F2, maximization): This is converted to a minimization objective F2' = 1 - F2, where F2 is the utilization rate of the reservoir's post-flood water storage. The formula is: (10) Where: V reservoir V represents the actual water storage capacity of the reservoir after the flood season (m³). full V represents the total reservoir capacity (m³). dead The dead storage capacity of the reservoir is (m³).

[0081] ③ Engineering safety objective (F3, minimization): Quantify the duration of gate over-opening and the number of pump station overload operations, using the following formula: (11) Among them, T overopen-k Let N be the over-opening duration (h) of the Kth gate, where K is the total number of gates; overload-l N represents the number of overload operations for the l-th pump station. total L represents the total number of pump station operations, and L represents the total number of pump stations.

[0082] 2) Specific construction formula of the penalty function for water conservancy scheduling A nonlinear penalty function P(X) is constructed to quantify the physical quantities of engineering constraints into penalty values, which are then embedded into the objective function to achieve hard processing of the constraints. The fusion formula between the penalty function and the objective function is as follows: F total = F + P(X) (12) Where F is the original three-objective optimization function, P(X) is the penalty function, P(X) = 0 when the engineering parameters meet the constraints; when the constraints are violated, P(X) is a non-linearly increasing penalty value, the specific formula of which is: (13) The first part consists of flow / opening constraints and penalties (reservoir discharge, gate opening), X m Let X be the actual parameter value of the m-th project. m-max The upper limit of the parameters for the m-th project is determined by the squared term, which implements a non-linear increase in the penalty value, with a coefficient of 10 as the penalty weight; the second part is the penalty for time / frequency constraints (continuous operation time of the pumping station, duration of gate over-opening), T n T represents the actual duration / number of times for the nth project. n-max The maximum duration / number of attempts for the nth project is determined by the cube term, which implements a more severe penalty (to prevent the project from being overloaded for extended periods). A coefficient of 20 represents the penalty weight. The penalty function's value range is: P(X) ≥ 0; the more severe the constraint violation, the larger P(X) becomes. The optimization objective is F. total The larger the value, the more automatically the algorithm will discard the solution, thus achieving hard processing of engineering constraints.

[0083] (3) Step 3: Fast non-dominated sorting and crowding calculation The equal-weighted congestion calculation of the standard NSGA-II is improved by introducing target weights and constructing a weighted congestion formula to highlight the importance of flood control safety objectives. The formula is as follows: (14) Among them, C i Let ω be the weighted crowding degree of the i-th individual. j The weights for the j-th objective are (ω1 = 0.6, ω2 = 0.2, ω3 = 0.2); F j (i + 1), F j (i - 1) represents the objective function values ​​of the (i + 1)th and (i - 1)th individuals in the j-th objective; F j-max F j-min Let be the maximum and minimum function values ​​of the j-th objective.

[0084] (4) Step 4: Genetic evolution operation Genetic evolution operations are performed using adaptive crossover mutation probability, replacing the fixed crossover probability (P) of standard NSGA-II. c=0.8), mutation probability (P) m =0.05) is improved to an adaptive probability, which is dynamically adjusted according to the overall fitness of individuals in the population. The formula is: (15) in: , , where F is the upper and lower bounds of probability; i F represents the overall fitness of the current individual. max F min The maximum and minimum overall fitness of the population; individuals with higher fitness have a lower probability of crossover mutation, thus retaining superior individuals; individuals with lower fitness have a higher probability of crossover mutation, thus promoting population evolution.

[0085] (5) Step 5: Local search operation of hill climbing algorithm Search scope: For individuals after crossover mutation, search for a better solution within the parameter neighborhood (±10%); Search logic: Calculate the objective function value of each point in the neighborhood, retain the parameters corresponding to the optimal value, and improve the local convergence speed.

[0086] (6) Step 6: Iteration termination judgment Termination condition: Number of iterations G ≥ Maximum number of iterations G max (1000 times), or the overall fitness change rate ≤10 for 20 consecutive generations. ; Solution selection: Output the Pareto optimal solution set (about 10-15 solutions), provide 3-5 candidate solutions based on watershed needs (prioritize flood control during the flood season and prioritize water storage at the end of the flood season), and support manual adjustment of weights for further optimization.

[0087] Step S5: Output the optimal scheduling scheme set and the coupled flood simulation results to the interactive interface for visualization.

[0088] Specifically, the optimized scheduling scheme and coupled simulation results are output in the watershed joint scheduling hydro-hydrodynamic coupled flood simulation method of this embodiment, as follows: Output content: Engineering dispatch instructions (reservoir discharge time curve, flood storage and detention area activation time / flood diversion volume, gate opening time table, pump station start and stop time / flow); simulation results (flood evolution animation, inundation area map, engineering operation status curve); visualization implementation: based on OpenGL + rendering engine, frame rate ≥25f ps, supports zooming, panning, and highlighting of inundated areas; result export: supports export in SHP (inundation area), TIFF (water level distribution), and Excel (dispatch parameters) formats, compatible with GIS platforms.

[0089] Therefore, the watershed joint scheduling hydro-hydrodynamic coupled flood simulation method of the present invention has the following technical advantages: 1. Significantly improved real-time performance and accuracy of hydrological and hydrodynamic coupling: Through the piecewise linear mapping formula from water level to roughness, the improved Muskingan method of the Xin'anjiang model, and the standardized data exchange interface, quantitative feedback correction of coupling parameters was achieved, coupling lag was reduced by 50%, and the deviation between simulation results and actual flood processes was ≤5%; at the same time, GPU parallel acceleration improved the efficiency of coupling calculation by 4-6 times, meeting the requirements of real-time scheduling.

[0090] 2. The coordination and conflict resolution capabilities of multi-project joint scheduling are significantly enhanced: By using the project scheduling priority matrix, conflict resolution decision tree, step-by-step scheduling logic for extreme conditions, and dynamic calculation formula for linkage coefficient, the abstract problem of "automatic adjustment" in the original technology has been solved, realizing the quantification and precision of multi-project scheduling. The matching degree between the scheduling scheme and the actual flood control needs has been improved by more than 30%, effectively avoiding project scheduling conflicts.

[0091] 3. Significantly improved creativity and engineering feasibility of multi-objective optimization: Through five substantial improvements to the NSGA-II algorithm, the construction of a special penalty function for water conservancy scheduling, and the quantitative calculation of the objective function, the optimization results, which are different from textbook-level standard algorithms, not only meet the core requirements of flood control safety, but also take into account water resource utilization and engineering safety. The reservoir storage capacity at the end of the flood season is increased by 10%-15%, the gate over-opening time is reduced by 80%, and the number of pump station overload operations is reduced by 90%.

[0092] 4. High engineering feasibility of GPU acceleration: Through explicit parallel strategies, mesh partitioning, and memory optimization methods, efficient parallel acceleration of solving the Saint-Venant equations is achieved, reducing the simulation time of complex watersheds to within 40 minutes. Moreover, this method can be ported to other hydraulic and hydrodynamic calculation scenarios, and has versatility. Example 2:

[0093] This invention provides a watershed joint scheduling hydrological and hydrodynamic coupled flood simulation system, see [link / reference]. Figure 5 As shown, the system 500 includes: Data module 501 is configured to collect multi-source watershed data and perform standardized preprocessing on the multi-source watershed data to obtain preprocessed watershed data. The coupled calculation module 502 is configured to use a hydrological and hydrodynamic model based on a two-way coupling mechanism to perform real-time coupled flood simulation calculations on preprocessed watershed data to obtain coupled flood simulation results. The joint scheduling module 503 is configured to process the coupled flood simulation results through a pre-set engineering scheduling priority matrix, a dynamic calculation formula for linkage coefficients, a conflict resolution decision tree, extreme working condition scheduling steps, and global collaborative constraints, so as to output an initial scheduling scheme. The multi-objective optimization module 504 is configured to iteratively optimize the initial scheduling scheme based on the flood control safety objective, water resource utilization objective, and engineering safety objective, using the improved NSGA-II algorithm to output the optimal scheduling scheme set. Interactive module 505 is configured to output the optimal scheduling scheme set and coupled flood simulation results to the interactive interface for visualization.

[0094] It should be noted that the data module 501, coupled computing module 502, joint scheduling module 503, multi-objective optimization module 504, and interaction processing module 505 in this embodiment are merely exemplary names. They can also be named according to their functions at the level of the hardware system. For example, they can be named as: data layer, coupled computing, joint scheduling layer, multi-objective optimization layer, and interaction layer, respectively. No specific limitation is made here.

[0095] For details, see Figure 6 As shown, in this embodiment, the data module 5012 data layer of the watershed joint scheduling hydro-hydrodynamic coupled flood simulation system ( Figure 6 At the lowest level, it integrates a multi-source data access submodule (MQTT protocol interface, file import interface), a data preprocessing submodule (GDAL / PROJ tool, outlier handling algorithm), an SDTS format standardization submodule (to convert multi-source data to the water conservancy industry standard format), and a data storage submodule (SQLite3 database), realizing the integration of "access-processing-storage"; the standardized dataset is output to the upper layer through the interface, supporting data traceability (querying the entire process of collection, processing, and use by data identifier).

[0096] Coupled computing module 502, i.e., coupled computing layer ( Figure 6 The second layer includes a hydrological calculation engine (improved core code of the Xin'anjiang model, integrating a dynamic confluence correction module), a hydrodynamic calculation engine (Saint-Venant equation solver), and a real-time coupling engine (30-minute bidirectional feedback interface, fuzzy adaptive correction algorithm). The hydrological engine outputs to the hydrodynamic engine, which corrects the hydrological engine parameters through feedback from the coupling engine. It also includes a standardized data exchange interface submodule and a GPU parallel computing engine (integrating grid partitioning and memory optimization submodules) to achieve real-time coupling and efficient calculation of hydrological and hydrodynamic parameters.

[0097] Joint scheduling module 503, i.e., joint scheduling layer ( Figure 6The third layer consists of a reservoir scheduling submodule, a flood storage and detention area scheduling submodule, a gate scheduling submodule, a pumping station scheduling submodule, an engineering priority determination submodule, a conflict resolution decision tree submodule, an extreme condition scheduling submodule, a linkage coefficient dynamic calculation submodule, and a collaborative constraint submodule, which realizes the collaborative scheduling and conflict resolution of multiple projects. Each scheduling submodule executes scheduling logic based on the coupled calculation results, outputs an initial scheduling plan, and realizes the linkage scheduling of multiple projects through standardized interface calls.

[0098] Multi-objective optimization module 504, i.e., multi-objective optimization layer ( Figure 1 The fourth layer includes a target function construction submodule (three-objective weighted calculation), an improved NSGA-II algorithm engine (integrating Latin hypercube sampling, adaptive crossover mutation, and weighted crowding calculation submodules), and a penalty function construction submodule, which realizes intelligent optimization of multiple objectives. It takes an initial scheduling scheme as input, outputs the optimized scheme, and supports multi-objective parameter configuration and optimal scheme output.

[0099] Interaction module 505, i.e., the interaction layer ( Figure 1 The top layer is a cross-platform GUI interface built on QT, providing functions such as data upload (interfacing with the data layer), parameter configuration (interfacing with the optimization layer), calculation monitoring (interfacing with the coupled calculation layer + scheduling layer), extreme working condition early warning, result viewing (interfacing with the optimization layer), and report export. The dynamic visualization sub-module is based on OpenGL to render simulation results and is compatible with Windows, Linux and domestic operating systems (Kylin, Tongxin). Example 3:

[0100] This invention discloses an electronic device. The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the watershed joint scheduling hydro-hydraulic coupling flood simulation method according to any one of the embodiments disclosed in this invention.

[0101] Figure 7 This is a structural diagram of an electronic device according to an embodiment of the present invention, such as... Figure 7As shown, the electronic device includes a processor, memory, communication interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, Near Field Communication (NFC), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the device's casing, or an external keyboard, touchpad, or mouse.

[0102] Those skilled in the art will understand that Figure 7 The structure shown is merely a structural diagram of the part related to the technical solution of this disclosure and does not constitute a limitation on the electronic device to which the solution of this application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements. Example 4:

[0103] This invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a watershed joint scheduling hydro-hydraulic coupling flood simulation method according to any one of the embodiments of this invention.

[0104] Please note that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments have been described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification. The above embodiments only illustrate several implementation methods of this application, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be pointed out that for those skilled in the art, several modifications and improvements can be made without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

[0105] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0106] While specific embodiments of the invention have been described in detail by way of examples, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of the invention. Those skilled in the art should understand that modifications can be made to the above embodiments without departing from the scope and spirit of the invention. The scope of the invention is defined by the appended claims.

Claims

1. A method for simulating floods by coupling hydrological and hydrodynamic factors in watershed joint scheduling, characterized in that, Place The methods include: Step S1: Collect multi-source watershed data and perform standardized preprocessing on the multi-source watershed data to obtain preprocessed watershed data; Step S2: Use a hydrological and hydrodynamic model based on a two-way coupling mechanism to perform real-time coupled flood simulation calculations on the preprocessed watershed data to obtain coupled flood simulation results; Step S3: Process the coupled flood simulation results using a pre-set engineering scheduling priority matrix, dynamic calculation formula for linkage coefficients, conflict resolution decision tree, extreme working condition scheduling steps, and global collaborative constraints to output an initial scheduling scheme; Step S4: Based on the flood control safety objective, water resource utilization objective, and engineering safety objective, the initial scheduling scheme is iteratively optimized using the improved NSGA-II algorithm to output the optimal scheduling scheme set; Step S5: Output the optimal scheduling scheme set and the coupled flood simulation results to the interactive interface for visualization.

2. The watershed joint scheduling hydro-hydrodynamic coupled flood simulation method according to claim 1, characterized in that, In step S1, the multi-source watershed data includes: basic data, real-time data, and scheduling rule data; The basic data specifically includes: DEM topographic data, land use data, river cross-section data, and flood control engineering parameter data; the real-time data specifically includes: real-time data from rainfall stations, real-time data from water level stations, and real-time data from flow stations; the scheduling rule data specifically includes: reservoir flood control scheduling curve data, flood storage and detention area flood control scheduling plan data, gate graded control rule data, and pump station linkage drainage logic rule data.

3. The watershed joint scheduling hydro-hydrodynamic coupled flood simulation method according to claim 1, characterized in that, In step S2, the hydrological and hydrodynamic model includes a hydrological sub-model and a hydrodynamic sub-model; Step S2 specifically includes: Step S21: The hydrological sub-model uses the improved Xin'anjiang model to calculate the runoff generation and confluence process of the preprocessed watershed data to obtain the hydrological model calculation results that include the river inflow process; Step S22: The hydrodynamic sub-model is based on the two-dimensional Saint-Venant equations. The finite volume method is used to solve the hydrodynamic model calculation results and the topographic data contained in the preprocessed watershed data to obtain the hydrodynamic model calculation results containing real-time water level and velocity distribution. Step S23: Select key cross sections within the watershed as feedback nodes. Extract the real-time water level value and predicted water level value of the key cross section from the calculation results of the hydrodynamic model and the calculation results of the hydrological model, respectively. Calculate the cross section water level deviation between the real-time water level value and the predicted water level value of the key cross section. When the cross section water level deviation exceeds a preset threshold, correct the roughness parameter of the confluence link in the hydrological sub-model according to the preset mapping formula between water level deviation and roughness, so as to realize the dynamic update of the parameters of the hydrological sub-model.

4. The watershed joint scheduling hydro-hydrodynamic coupled flood simulation method according to claim 3, characterized in that, In step S21, the improved Xin'anjiang model specifically involves: using the improved Muskingum method for flow calculation to achieve the correlation mapping between roughness values ​​and Muskingum method parameters; The specific formula for the improved Muskingan method is expressed as follows: (1) in, The improved outflow rate at the end of the time period after roughness correction; This represents the inflow rate at the end of the time period; This represents the inflow rate at the beginning of the time period; The outflow rate at the beginning of the time period; , , The roughness-corrected Muskingen calculus coefficients are expressed as follows: In the formula For time step; These are the Muskingan method storage coefficient and flow rate specificity factor after roughness correction, respectively, which are linearly related to the roughness value n, and the correlation formula is: , , The initial storage capacity coefficient and initial flow rate weight factor are for the Muskingan method. The initial value of roughness, This is the corrected roughness value.

5. The watershed joint scheduling hydro-hydrodynamic coupled flood simulation method according to claim 3, characterized in that, In step S22, a multi-level parallel strategy of segmenting the river channel and dividing the two-dimensional grid is adopted to distribute the solution calculation task to the multi-stream processor of the GPU for processing, so as to achieve computational acceleration.

6. The watershed joint scheduling hydro-hydrodynamic coupled flood simulation method according to claim 3, characterized in that, In step S23, the roughness parameter of the confluence link in the hydrological sub-model is corrected according to the preset mapping formula between water level deviation and roughness, specifically as follows: The cross-sectional water level deviation is quantitatively converted into a roughness correction coefficient using a piecewise linear preset mapping formula. Then, the product of the roughness correction coefficient and the initial roughness value is calculated to obtain the roughness correction value. Based on the principle of nearest correction in sub-basins, the roughness parameters of the confluence link in the hydrological sub-model are corrected according to the roughness correction value, so as to realize the dynamic updating of the parameters of the hydrological sub-model. The piecewise linear preset mapping formula is expressed as follows: (2) In the formula, This is the roughness correction coefficient, derived from the cross-sectional water level deviation. The only certainty is that m is the unit of length, the meter.

7. The watershed joint scheduling hydro-hydrodynamic coupled flood simulation method according to claim 1, characterized in that, Step S4 specifically includes: Step S41: Initialize the population of scheduling parameters using the Latin hypercube sampling method. The scheduling parameters include the reservoir pre-discharge coefficient, flood diversion priority of flood storage and detention areas, gate opening adjustment coefficient, and pumping station drainage coefficient. Step S42: Construct a three-objective optimization function based on flood control safety objective, water resource utilization objective, and engineering safety objective, and calculate the total objective function value of each individual in the population by combining the water conservancy scheduling-specific nonlinear penalty function, wherein the water conservancy scheduling-specific nonlinear penalty function is used to quantify engineering constraints into penalty values; Step S43: Perform a fast non-dominated sort on the population, and calculate the weighted crowding degree for each individual using a weighted crowding degree calculation method, wherein the weighted crowding degree assigns a higher weight to the flood control safety target than to the water resource utilization target and the engineering safety target; Step S44: Perform genetic evolution operations on the population using adaptive crossover mutation probability, wherein the adaptive crossover mutation probability is dynamically adjusted according to the overall fitness of individuals in the population; Step S45: Perform a hill-climbing algorithm local search operation on each individual after the genetic evolution operation to search for a better solution in the parameter neighborhood; Step S46: Determine whether the iteration termination condition is met. If it is met, output the optimal scheduling optimal solution set. If it is not met, return to step S42 and repeat until the termination condition is met.

8. A watershed joint scheduling hydrological-hydraulic coupled flood simulation system, characterized in that, The system includes: The data module is configured to collect multi-source watershed data and perform standardized preprocessing on the multi-source watershed data to obtain preprocessed watershed data. The coupled calculation module is configured to use a hydrological and hydrodynamic model based on a two-way coupling mechanism to perform real-time coupled flood simulation calculations on the preprocessed watershed data to obtain coupled flood simulation results. The joint scheduling module is configured to process the coupled flood simulation results through a pre-set engineering scheduling priority matrix, a dynamic calculation formula for linkage coefficients, a conflict resolution decision tree, extreme working condition scheduling steps, and global collaborative constraints, so as to output an initial scheduling scheme. The multi-objective optimization module is configured to iteratively optimize the initial scheduling scheme based on flood control safety objectives, water resource utilization objectives, and engineering safety objectives using an improved NSGA-II algorithm, so as to output an optimal scheduling scheme set. The interactive module is configured to output the optimal scheduling scheme set and the coupled flood simulation results to the interactive interface for visualization.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, it implements the steps in the watershed joint scheduling hydrological and hydrodynamic coupled flood simulation method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the watershed joint scheduling hydrological-hydrodynamic coupled flood simulation method according to any one of claims 1 to 7.

Images