An ecological regulation method for relieving retention of the biogenic substance in a deep and large reservoir

Abstract

The application discloses an ecological regulation method for relieving the retention of the biogenic substance of a deep and large reservoir, and belongs to the technical field of water resource management and environmental protection technology. The steps are as follows: first, collecting characteristic parameters and multiple types of hydrological and water quality monitoring data of the deep and large reservoir; second, constructing a three-dimensional hydrodynamic-water quality model of the deep and large reservoir; third, based on the three-dimensional hydrodynamic-water quality model of the deep and large reservoir, combining a long series of multiple inputs and multiple output sequences, a data-driven proxy model is constructed; fourth, constructing a reservoir optimization regulation model to deduce suitable regulation rules for decision makers. The application improves the hydrodynamic condition by discharge to reduce the retention of the biogenic substance, and realizes the activation and transportation by using the water temperature stratification condition of the reservoir area, so that the effective biogenic substance is transported to the downstream while the power generation capacity is ensured, thereby realizing the 'quantity, energy and quality comprehensive regulation and control' considering the power generation, ecological flow and biogenic substance, and providing technical support for the comprehensive management of optimizing the utilization of water resources and ecological environment protection of the basin.

Description

Technical Field

[0001] This invention belongs to the field of water resource management and environmental protection technology, and relates to an ecological scheduling method to alleviate the retention of biogenic substances in the Shenda Reservoir. Background Technology

[0002] The construction and operation of the Shenzhen-Dalian Reservoir have a significant impact on the water quality and ecosystem of the reservoir area and downstream rivers. Reservoir impoundment traps biogenic substances such as nitrogen and phosphorus, leading to the accumulation of total nitrogen (TN) and total phosphorus (TP) in the reservoir area, which may cause eutrophication in the long term. Simultaneously, downstream rivers are prone to oligotrophication due to the release of cold water and reduced nutrients. Therefore, developing scheduling strategies to alleviate the retention of biogenic substances in the Shenzhen-Dalian Reservoir is of great importance.

[0003] To address the aforementioned issues, Chinese invention patent CN111080157B proposes a method and system for scheduling phosphorus discharge from cascade hydropower stations. This method integrates the dual objectives of total power generation and phosphorus discharge from the cascade hydropower stations into a single scheduling objective, generates a sequence of scheduling objectives with different weight ratios, and then optimizes each objective to obtain the water level scheduling process, thereby mitigating the ecological and environmental problems caused by hydropower station construction. Chinese invention patent CN115659641B proposes a machine learning-based method for controlling algal blooms. This method mainly constructs an algal bloom control scheduling model that includes economic and ecological objective functions. The ecological objective function is established using an LSTM model based on the comprehensive nutrient index of the water body and phytoplankton density. The non-dominated solution to the multi-objective problem is obtained through the Cuckoo Optimization Algorithm, from which a satisfactory ecological scheduling scheme is selected. However, the water quality scheduling objectives of the above patents mainly rely on downstream flux or reservoir nitrogen and phosphorus concentrations and their derived nutrient indices, and use a fixed intake water level as the scheduling object. For the Shenzhen-Daqing Reservoir, the internal nitrogen and phosphorus migration and transformation processes exhibit a distinct three-stage characteristic: "retention-transformation-transportation." The dominant mechanisms differ significantly across different periods and spatial locations: during the non-flood season, the area in front of the dam is primarily characterized by the activation and release of dissolved nitrogen and phosphorus, while during the flood season, the central reservoir area is dominated by the retention and interception of particulate nitrogen and phosphorus. Current technologies do not fully consider the migration and transformation mechanisms of biomass in the Shenzhen-Daqing Reservoir, resulting in shortcomings in the scientific nature of decision-making and the precision of scheduling, leaving room for improvement.

[0004] In summary, how to combine the migration and transformation mechanisms of biological elements in the Shenzhen-Dalian Reservoir to provide an ecological scheduling method to alleviate the retention of biological materials in the reservoir, balance the contradiction between power generation and the environment, reduce eutrophication in the reservoir area, and ensure the water quality of downstream rivers has become an urgent problem to be solved. Summary of the Invention

[0005] To address the shortcomings of existing ecological scheduling methods that fail to fully consider the spatiotemporal evolution of biogenic substances within reservoirs, making it difficult to effectively alleviate nitrogen and phosphorus retention while ensuring power generation, this invention provides an ecological scheduling method for mitigating biogenic substance retention in deep and large reservoirs. This invention improves hydrodynamic conditions by releasing water to reduce biogenic substance retention and utilizes the stratified water temperature in the reservoir area to activate and transport biogenic substances. While ensuring power generation, it strengthens the downstream transport of effective biogenic substances, thereby achieving comprehensive "quantitative, energy-quality regulation" that considers power generation, ecological flow, and biogenic substances. This provides technical support for optimizing the integrated management of watershed water resource utilization and ecological environmental protection.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] An ecological regulation method for alleviating the retention of biogenic materials in the Shenda Reservoir includes the following steps:

[0008] The first step involves collecting characteristic parameters of large and deep reservoirs in the study basin, as well as various types of hydrological and water quality monitoring data, as detailed below:

[0009] The characteristic parameters of the deep reservoir mainly include: digital elevation model (DEM), dam height, total reservoir capacity, normal water level, and water level-capacity curve.

[0010] The various types of hydrological and water quality monitoring data mainly include: meteorological data, hydrological data, pollution load data, inflow data, outflow data, water level data, surface water temperature data, vertical water temperature data, surface water quality data, vertical water quality profile data, and point source and non-point source pollution data.

[0011] The second step is to construct a three-dimensional hydrodynamic-water quality model of the Shenzhen Reservoir. Based on the characteristic parameters of the Shenzhen Reservoir and various types of hydrological and water quality monitoring data collected in the first step, a three-dimensional hydrodynamic module and a water quality module of the Shenzhen Reservoir are constructed, and their coupling is calibrated to form a reliable three-dimensional hydrodynamic-water quality model of the Shenzhen Reservoir, as detailed below:

[0012] Step 2.1: Based on the characteristic parameters of the Shenda Reservoir and various hydrological and water quality monitoring data collected in Step 1, the planar and vertical grids are divided, topographic and water depth data are input, and boundary conditions are set to construct the three-dimensional hydrodynamic module of the Shenda Reservoir; specifically:

[0013] Grid type selection: Curved coordinate grids are used for planar grid division, effectively conforming to the irregular and curved reservoir boundary. Grid resolution is increased in areas with significant river channel curvature, while moderate simplification is maintained in areas with smooth boundaries. The σ-coordinate system is selected for vertical grid division to better reflect the complex underwater topography and accurately represent water depth variations.

[0014] Topographic and water depth data input: Based on the spatial data of the Digital Elevation Model (DEM), the elevation data of the entire reservoir area is extracted, and each grid cell in the reservoir area is assigned an accurate elevation value to reflect the topographic undulations and provide a basis for simulating water flow, material transport and water quality changes.

[0015] Boundary condition settings: Setting boundary conditions mainly includes input data, initial background value settings, and warm-up period selection. For the inflow boundary, physical elements such as flow rate and temperature, meteorological elements such as precipitation and evaporation, and water quality load from upstream are required. The model's initial conditions are set based on key elements such as flow rate, water temperature, meteorological conditions, and water quality corresponding to the first day of the simulation period. A hot start method is used for initialization, with the output at the end of the first year of the simulation period serving as the hot start input for the following year.

[0016] Thus, a three-dimensional hydrodynamic module for the deep reservoir was constructed. To further optimize computational performance, efficient computing functions such as parallel computing, nested mesh, and computational domain decomposition were adopted to enable it to perform large-scale hydrodynamic simulations and achieve local fine-grained calculations in key areas, thereby improving its adaptability and simulation accuracy in complex hydrodynamic environments.

[0017] Step 2.2 involves selecting water quality-related processes and setting key parameters to construct a water quality module. Specifically:

[0018] Selection of water quality-related processes: comprehensively consider the migration and transformation processes of key water quality components such as dissolved oxygen (DO), ammonia nitrogen (NH4), nitrate nitrogen (NO3), orthophosphate (PO4), particulate nitrogen (DetN), and particulate phosphorus (DetP) in water bodies.

[0019] Key parameters were set: gravitational acceleration, water density, and air density; corresponding drag coefficients and wind speed thresholds were set for different wind speed ranges; and related parameters such as friction coefficient, horizontal eddy viscosity, and eddy diffusion coefficient were also included. Regarding heat exchange at the water-air interface, Secchi depth, Dalton number, and Stanton number were key parameters controlling water surface evaporation and heat exchange processes. For water quality parameters, parameters such as nitrogen and phosphorus decomposition rate constants and sedimentation velocities were set to simulate the conversion of particulate nitrogen and phosphorus to dissolved nitrogen and phosphorus, playing a crucial role in simulating nitrogen and phosphorus migration, sedimentation, and re-release.

[0020] Thus, a water quality module was constructed. This module uses the finite volume method for numerical solution. Its numerical solver can efficiently handle convection, diffusion and reaction equations, accurately calculate water transport, pollutant diffusion and biogeochemical processes, ensure the mass conservation of each substance during numerical calculation, and improve the accuracy and stability of the simulation.

[0021] Step 2.3: Couple the water quality module with the hydrodynamic module to form a three-dimensional hydrodynamic-water quality model of the deep reservoir, giving it comprehensive water quality processes and efficient computing capabilities.

[0022] Coupling Mechanism: The three-dimensional hydrodynamic module and the water quality module of the deep reservoir adopt a loose coupling mechanism to achieve collaborative operation. The basic principle is to use the hydrodynamic process as the external driving force for water quality evolution, and drive the transport and transformation of materials in the water body through dynamic flow field information, thereby realizing the dynamic simulation of material migration and change in complex water bodies.

[0023] Coupling steps: First, the hydrodynamic module simulates and outputs key hydrodynamic parameters such as flow velocity, water depth, temperature, and turbulent diffusion coefficient. These parameters vary with time and space, reflecting the dynamic characteristics of real hydrological processes. Then, these parameters are passed to the water quality module as forced inputs in each spatial grid and at each time step, driving it to dynamically simulate the convection, diffusion, and reaction processes of matter.

[0024] Step 2.4: Based on the characteristic parameters of the Shenda Reservoir and various hydrological and water quality monitoring data collected in the first step, the three-dimensional hydrodynamic-water quality model of the Shenda Reservoir is calibrated and verified in an integrated manner, encompassing multiple processes, multiple elements, multiple time and space, and multiple indicators.

[0025] In terms of "multi-processes": the focus is on the dynamic linkage and physical consistency of key processes such as hydrodynamic evolution, water temperature structure, and water quality transformation.

[0026] In terms of "multi-factors": multiple variables such as water level, flow rate, water temperature, total nitrogen (TN), total phosphorus (TP), dissolved oxygen (DO), and biochemical oxygen demand (BOD) are included to improve the comprehensiveness and accuracy of parameter constraints.

[0027] In terms of "multi-temporal and spatial aspects": it covers high-frequency water level and water temperature data at the daily scale throughout the year, as well as key periods such as typical seasonal scales and flood season and non-flood season, and conducts multi-time period, multi-section, and multi-depth collaborative verification at key locations such as the main stream, tributaries, reservoir, and dam front, as well as vertically.

[0028] Regarding "multiple indicators": Avoid relying solely on a single or few indicators to judge model performance. Instead, comprehensively evaluate the performance and physical consistency of the three-dimensional hydrodynamic-water quality model of the Shenzhen Reservoir under different processes, elements, and spatiotemporal scales by using multiple indicators such as the Nash-Sutcliffe efficiency coefficient (NSE), coefficient of determination (R²), root mean square error (RMSE), and absolute relative deviation (ADR%).

[0029] Step 2.5: Based on the calibrated and verified three-dimensional hydrodynamic-water quality model of the Shenzhen Reservoir, organize its input and simulation output results to form a training dataset for constructing the subsequent data-driven surrogate model.

[0030] The third step involves constructing a data-driven surrogate model based on the three-dimensional hydrodynamic-water quality model of the Shenda Reservoir built in the second step, as detailed below:

[0031] Step 3.1, Organize and divide the training dataset:

[0032] Based on step 2.5, the training dataset for the data-driven surrogate model is obtained. To ensure the breadth and representativeness of the surrogate model training, the training dataset is divided into a training set, a validation set, and a test set, with proportions of 60%, 20%, and 20%, respectively.

[0033] Step 3.2, Building the Proxy Model. After preparing the training dataset in Step 3.1, we proceed to the proxy model building phase. The detailed steps are as follows:

[0034] Step 3.2.1, Proxy Model Architecture Design:

[0035] Input layer design: The input layer of the surrogate model receives multiple long sequence features, including long series of hydrological and water quality data input to the three-dimensional hydrodynamic-water quality model of the large reservoir, as well as multi-layer water quality concentration data output by the three-dimensional hydrodynamic-water quality model of the large reservoir; if the dimension of the input features is n, then the input layer contains n neurons.

[0036] Hidden layer design: To capture the complex nonlinear relationship between water quality and environmental drivers, the model contains multiple hidden layers and uses ReLU (Rectified Linear Unit) as the activation function. ReLU can effectively alleviate the gradient vanishing problem and improve the convergence speed.

[0037] Output layer design: The model's output layer consists of multiple neurons used to predict water quality-related indicators. Single-point prediction is employed, meaning the output shows the water quality variation over time at a representative location or layer. Since water quality prediction is a regression problem, the output layer uses a linear activation function, suitable for numerical prediction.

[0038] Step 3.2.2, Training Process:

[0039] Training dataset: Using the training set from step 3.1, the input features are the meteorological data, hydrological data, and pollution load data collected in step 1. The target variable is the water quality change process (e.g., surface water quality, vertical water quality, etc.). The loss function is the mean squared error (MSE), which calculates the difference between the predicted values ​​of the surrogate model and the output values ​​of the three-dimensional hydrodynamic-water quality model of the Shenda Reservoir.

[0040] (1)

[0041] in, For the first The predicted value for each sample, These are actual observations. The total number of samples, This indicates the sample index.

[0042] Proxy model performance optimization: The Adam optimizer is employed. This optimization algorithm combines adaptive learning rate and momentum methods to improve the efficiency and convergence speed of surrogate model training. The Adam algorithm updates the parameters of each layer according to the following update rules:

[0043] (2)

[0044] in, For learning rate, and For estimates of the first and second moments, To prevent division by zero of constants; For the first Model parameters obtained through time-period iteration; For the first Model parameters obtained through time-period iteration.

[0045] Regularization: To avoid overfitting, regularization techniques, also known as weight decay, are used. This involves adding a regularization term to the loss function to control model complexity.

[0046] (3)

[0047] in, The regularization coefficient is . These are the weight parameters in the model; This represents the total loss.

[0048] Training strategy: An early stopping strategy is employed to prevent overfitting. Training is stopped early when performance on the validation set no longer improves. This strategy effectively saves training time while avoiding overfitting to the training data.

[0049] Step 3.3, Model Validation and Evaluation:

[0050] To ensure the accuracy and generalization ability of the model, the surrogate model underwent various validations and evaluations. Cross-validation was primarily used to ensure the stability and reliability of the surrogate model. Specifically:

[0051] Cross-validation: This method uses K-fold cross-validation to evaluate the model. The data is divided into K subsets, and one subset is used as the validation set each time, while the remaining subsets are used for training. Finally, the average performance of the surrogate model is evaluated. This method effectively reduces the random errors introduced by data partitioning.

[0052] Evaluation metrics: The main metrics used are root mean square error (RMSE) and Nash efficiency coefficient (NSE). The closer the value is to 1, the better the model's performance.

[0053] Step 4: Construct a reservoir optimal scheduling model and couple the data-driven surrogate model obtained in Step 3 with the reservoir optimal scheduling model as a water quality response prediction tool. This aims to satisfy multiple water quality objectives while considering power generation needs, and to derive appropriate scheduling rules for decision-makers. Specifically:

[0054] Step 4.1, Define decision variables:

[0055] Determining Decision Variables: To ensure that predetermined ecologically relevant water quality and power generation targets can be achieved under different inflow scenarios, the reservoir scheduling process needs to be treated as a decision variable, making it applicable under uncertainty. The scheduling rules consist of a combination of target reservoir capacity, output coefficient, and discharge location set for each scheduling period, used to determine the correspondence between unit output and reservoir capacity.

[0056] Step 4.2, Set optimization goals:

[0057] Determine the power generation target: Maximize the multi-year average power generation of the Shenda Reservoir as the power generation efficiency optimization target, and provide an economic evaluation basis for subsequent dispatching schemes.

[0058] (4)

[0059] in, The total number of scheduling periods. Indicates the calculation period; To contribute to the power station This represents the average annual power generation over many years. For time step.

[0060] Determining downstream water quality targets: Considering that the main release of dissolved nitrogen and phosphorus in the Shenda Reservoir area during the non-flood season, and the main retention and interception of particulate nitrogen and phosphorus during the flood season, different water quality targets are set for different time periods to alleviate eutrophication in the reservoir area and oligotrophication downstream. During the non-flood season, the control target is mainly the concentration of dissolved nitrogen and phosphorus in the reservoir water, controlling their concentration as low as possible to improve the water body's self-purification capacity and reduce the risk of eutrophication in the reservoir area. During the flood season, the control target is mainly the downstream flux of biogenic nitrogen and phosphorus, optimizing the discharge strategy to achieve the transport of as much effective biogenic material as possible downstream, thereby alleviating the oligotrophication phenomenon in the downstream water body.

[0061] (5)

[0062] (6)

[0063] (7)

[0064] (8)

[0065] in, For the reservoir in The flow rate at any given moment, The layer where the reservoir discharges water is located. Time period TP concentration within, The layer where the reservoir discharges water is located. Time period TN concentration within, The average TN concentration across multiple layers in the reservoir area; CS TP The average TP concentration across multiple layers in the reservoir area; N represents the total number of monitoring times, and n represents the index of the number of monitoring times. The TP flux through the reservoir discharge section; TN flux through the reservoir discharge section.

[0066] Step 4.3, Set constraints:

[0067] Constraints: Optimal reservoir scheduling generally follows constraints such as water balance, reservoir capacity, reservoir discharge, power generation flow, reservoir output, and the range of scheduling rule parameters.

[0068] 1) Reservoir water balance constraints:

[0069] (9)

[0070] 2) Reservoir capacity constraints:

[0071] (10)

[0072] 3) Power output constraints of hydropower stations:

[0073] (11)

[0074] 4) Ecological flow constraints in river channels:

[0075] (12)

[0076] In the formula: For the reservoir in The water storage capacity at any given time, S t-1 Let be the water storage volume of the reservoir at time t-1; Dead storage capacity of the reservoir; This is the normal water storage capacity of the reservoir; For reservoir Inbound traffic at any given time It refers to the installed capacity of the reservoir, in MW; It is the ecological flow of the river downstream of the reservoir.

[0077] Step 4.4, Define the solution algorithm:

[0078] Optimization algorithm selection: Considering that the reservoir optimization scheduling model is multi-objective, the heuristic algorithm NSGA-III, which balances computational efficiency and solution quality, is selected to solve it.

[0079] By solving a multi-objective reservoir optimization scheduling model, a set of scheduling rules that meet power generation needs and water quality requirements were obtained, providing a scientific basis and technical support for the comprehensive management of water resources and water environment to alleviate the retention of biomass in the reservoir area and improve the water quality of downstream areas.

[0080] An ecological scheduling method for alleviating biomass retention in the Shenzhen-Dalian Reservoir is implemented based on the Shenzhen-Dalian Reservoir ecological scheduling system. This system includes a data acquisition module, a Shenzhen-Dalian Reservoir hydrodynamic-water quality model construction module, a data-driven proxy model construction module, and a reservoir scheduling optimization module. Specifically:

[0081] Data acquisition module: used to acquire characteristic parameters of the deep and large reservoirs in the research watershed and various types of hydrological and water quality monitoring data, providing basic input for hydrodynamic and water quality simulation of the deep and large reservoirs.

[0082] The three-dimensional hydrodynamic-water quality model construction module of Shenzhen Reservoir: Constructs a three-dimensional hydrodynamic module and a water quality module for the Shenzhen Reservoir to study the reservoir. By solving the continuity equation, momentum equation and water quality component transport equation, it simulates the spatiotemporal distribution characteristics of water quality indicators of the reservoir under historical operating conditions, and couples the two to form a reliable three-dimensional hydrodynamic-water quality model of Shenzhen Reservoir.

[0083] Data-driven surrogate model building module: Based on the simulation results of the three-dimensional hydrodynamic-water quality model of Shenzhen Reservoir and related input features and output results, a surrogate model suitable for water quality prediction is built;

[0084] Reservoir scheduling optimization module: Coupled with a data-driven proxy model, a reservoir optimization scheduling model is constructed, and a set of scheduling rules is derived under the premise of meeting power generation demand and water quality requirements.

[0085] The beneficial effects of this invention are as follows:

[0086] (1) This invention, by combining the migration and transformation mechanism of biogenic materials in the Shenzhen-Daqing Reservoir and setting differentiated water quality targets in different time periods, constructs a refined ecological scheduling strategy that integrates discharge flow, discharge time, and discharge location, and proposes an ecological scheduling technology to alleviate the retention of biogenic materials. This technology overcomes the shortcomings of existing technologies in terms of decision-making scientificity and scheduling refinement, and provides scientific basis and technical support for the operation of the Shenzhen-Daqing Reservoir.

[0087] (2) This invention uses a data-driven proxy model to perform a high-fidelity approximation of the hydrodynamic-water quality model of the Shenda Reservoir, thereby achieving rapid prediction of multi-level water quality in the reservoir area. This method overcomes the inefficiency caused by the complexity of calculation and long running time of traditional physical hydrodynamic-water quality models in practical applications, and improves the real-time performance and operability of reservoir ecological scheduling.

[0088] (3) This invention provides an efficient and intelligent decision support tool for ecological scheduling to alleviate the retention of biological materials in the deep reservoir. It not only improves the scientific and refined level of management, but also promotes the protection of the aquatic ecological environment and the sustainable use of water energy resources. Attached Figure Description

[0089] Figure 1 A flowchart of the ecological scheduling method for alleviating the retention of biogenic materials in the Shenda Reservoir provided by the present invention;

[0090] Figure 2 A schematic diagram of the structure of the deep reservoir hydrodynamic-water quality coupling model provided by the present invention;

[0091] Figure 3 Performance evaluation diagram of the data-driven proxy model provided by this invention. Detailed Implementation

[0092] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0093] The purpose of this invention is to provide an ecological scheduling method and system for alleviating the retention of biomass in the Shenda Reservoir. This system can balance the benefits of hydropower generation with the water quality control needs of the reservoir area and downstream areas, and scientifically and efficiently derive ecological scheduling strategies for the Shenda Reservoir, providing decision support for regional sustainable development and ecological protection.

[0094] To make the above-mentioned objectives, features, and advantages of the present invention more apparent and understandable, the Xiaowan Reservoir in the Lancang River Basin is used as an example below. The present invention will be further described in detail with reference to the accompanying drawings and specific embodiments, so as to promote the extension of the content of this patent to the Yarlung Tsangpo River Basin and other deep and large plateau reservoirs.

[0095] See Figure 1 Ecological regulation methods to alleviate the retention of biomass in the Shenda Reservoir include:

[0096] Step 1: Collect characteristic parameters and various hydrological and water quality monitoring data of the Shenda Reservoir, as detailed below:

[0097] The key characteristic parameters of the Shenda Reservoir include: elevation model (DEM), dam height, total reservoir capacity, normal water level, and water level-capacity curve. The Xiaowan Reservoir dam is 292 meters high, with a total reservoir capacity of 151.32 × 10⁻⁶ meters. 8 cubic meters. The normal water level of Xiaowan Reservoir is 1240 meters, and the corresponding water surface area is approximately 189 square kilometers.

[0098] The various types of hydrological and water quality monitoring data mainly include: meteorological data, inflow data, outflow data, water level data, surface water temperature data, vertical water temperature data, surface water quality data, vertical water quality profile data, and point source and non-point source pollution data.

[0099] The second step is to construct a three-dimensional hydrodynamic-water quality model of the Shenda Reservoir. Based on the characteristic parameters of the Shenda Reservoir and various hydrological and water quality monitoring data from the first step, a three-dimensional hydrodynamic module and a water quality module of the Shenda Reservoir are constructed, and their coupling is calibrated to form a reliable three-dimensional hydrodynamic-water quality model of the Shenda Reservoir. Figure 2 The details are as follows:

[0100] Step 2.1: Based on the characteristic parameters of the deep reservoir and the collection of various hydrological and water quality monitoring data from Step 1, the planar and vertical grids are divided, topographic and water depth data are input, and boundary conditions are set to construct the hydrodynamic module. Specifically:

[0101] Grid type selection: Curved coordinate grids are used for planar grid division, effectively conforming to the irregular and curved reservoir boundary. Grid resolution is increased in areas with significant river channel curvature, while moderate simplification is maintained in areas with smooth boundaries. The σ-coordinate system is selected for vertical grid division to better reflect the complex underwater topography and accurately represent water depth variations.

[0102] Topographic and water depth data input: Based on the spatial data of the digital elevation model (DEM), the elevation data of the entire reservoir area was extracted, and each grid cell in the reservoir area was assigned an accurate elevation value to reflect the topographic undulations and provide a basis for simulating water flow, material transport and water quality changes.

[0103] Boundary condition settings: Setting boundary conditions mainly includes input data, initial background value settings, and warm-up period selection. For the inflow boundary, physical elements such as flow rate and temperature, meteorological elements such as precipitation and evaporation, and water quality load from upstream are required. The model's initial conditions are set based on key elements such as flow rate, water temperature, meteorological conditions, and water quality corresponding to the first day of the simulation period. A hot start method is used for initialization, with the output at the end of the first year of the simulation period serving as the hot start input for the following year.

[0104] Thus, a hydrodynamic module was constructed. To further optimize computational performance, efficient computing functions such as parallel computing, nested meshes, and computational domain decomposition were adopted, enabling it to perform large-scale hydrodynamic simulations and achieve localized refined calculations in key areas, thereby improving its adaptability and simulation accuracy in complex hydrodynamic environments.

[0105] Step 2.2 involves the rational configuration of water quality-related processes and the setting of key parameters to construct a water quality module. Specifically:

[0106] Selection of water quality-related processes: The migration and transformation processes of key water quality components such as dissolved oxygen (DO), ammonia nitrogen (NH4), nitrate nitrogen (NO3), orthophosphate (PO4), particulate nitrogen (DetN), and particulate phosphorus (DetP) in water bodies were comprehensively considered.

[0107] Key parameter settings: Gravitational acceleration 9.81 m / s² 2 Water density 1000 kg / m³ 3 air density 1kg / m³ 3 Corresponding drag coefficients and wind speed thresholds are set for different wind speed ranges, with drag coefficient A being 0.00063, drag coefficient B being 0.00723, and drag coefficient C being 0.00723; the friction coefficient is 0.022, and the horizontal eddy viscosity is 2m. 2 / s and vortex diffusion coefficient 2m 2 Parameters such as / s were used. Regarding heat exchange at the water-air interface, a Secchi depth of 2 m, a Dalton number of 0.0013, and a Stanton number of 0.0013 were key parameters controlling water surface evaporation and heat exchange. For water quality parameters, nitrogen and phosphorus decomposition rate constants and settling velocities were set, with a settling velocity of 0.1 m / d and a nitrogen decomposition rate constant of 0.0003 d / s. -1 P decomposition rate constant d -1 .

[0108] Thus, a water quality module was constructed. This module uses the finite volume method for numerical solution. Its numerical solver can efficiently handle convection, diffusion and reaction equations, accurately calculate water transport, pollutant diffusion and biogeochemical processes, ensure the mass conservation of each substance during numerical calculation, and improve the accuracy and stability of the simulation.

[0109] Step 2.3: Couple the water quality module with the hydrodynamic module to form a three-dimensional hydrodynamic-water quality model of the deep reservoir, giving it comprehensive water quality processes and efficient computing capabilities.

[0110] Coupling Mechanism: The hydrodynamic module and the water quality module adopt a loose coupling mechanism to achieve collaborative operation. The basic principle is to use the hydrodynamic process as the external driving force for water quality evolution, and drive the transport and transformation of substances in the water body through dynamic flow field information, thereby realizing the dynamic simulation of material migration and change in complex water bodies.

[0111] Coupling steps: First, the hydrodynamic module simulates and outputs key hydrodynamic parameters such as flow velocity, water depth, temperature, and turbulent diffusion coefficient. These parameters vary with time and space, reflecting the dynamic characteristics of real hydrological processes. Then, these parameters are passed to the water quality module as forced inputs in each spatial grid and at each time step, driving it to dynamically simulate the convection, diffusion, and reaction processes of matter.

[0112] Step 2.4: Based on the multi-type hydrological and water quality monitoring data collected in Step 1, the three-dimensional hydrodynamic-water quality model of Shenzhen Reservoir is calibrated and verified in an integrated manner, encompassing multiple processes, multiple elements, multiple time and space, and multiple indicators.

[0113] In terms of "multi-processes": the focus is on the dynamic linkage and physical consistency of key processes such as hydrodynamic evolution, water temperature structure, and water quality transformation.

[0114] In terms of "multi-factors": multiple variables such as water level, flow rate, water temperature, total nitrogen (TN), total phosphorus (TP), dissolved oxygen (DO), and biochemical oxygen demand (BOD) are included to improve the comprehensiveness and accuracy of parameter constraints.

[0115] In terms of "multi-temporal and spatial" data: high-frequency water level and temperature data at the daily scale throughout the year, as well as key periods such as typical seasonal scales, flood season and non-flood season, and multi-time period, multi-section, and multi-depth collaborative verification in key locations such as the main stream, tributaries, reservoir, and dam front, and within a vertical depth of 150m.

[0116] Regarding the "multi-indicator" approach: To avoid relying solely on a single or few indicators to judge model performance, this invention comprehensively evaluates the model's performance and physical consistency across different processes, elements, and spatiotemporal scales by using a combination of indicators, including the Nash-Sutcliffe efficiency coefficient (NSE), coefficient of determination (R²), root mean square error (RMSE), and absolute relative deviation (ADR%). For water level, flow rate, and water temperature, the NSE is greater than 0.9, R² is greater than 0.95, RMSE is less than 1.2, and ADR% is less than 6.5%. For water quality elements, the standard for model performance evaluation is slightly lower. Existing research indicates that NSE > 0.3 has consistently been the accepted accuracy standard for most water quality model studies. Except for a few points where the NSE, R², RMSE, and ADR% for water quality elements generally meet the above requirements.

[0117] Step 2.5: Based on the calibrated and verified three-dimensional hydrodynamic-water quality model of the Shenzhen Reservoir, organize its input and simulation output results to form a training dataset for constructing the subsequent data-driven surrogate model.

[0118] The third step involves constructing a data-driven surrogate model based on the three-dimensional hydrodynamic-water quality model of the deep reservoir built in the second step, combined with its long series of multi-input and multi-output sequences. The details are as follows:

[0119] Step 3.1, Organize and divide the training dataset:

[0120] Training Dataset Partitioning: Based on step 2.5, the training dataset for the surrogate model was obtained. To ensure the broadness and representativeness of the surrogate model training, the dataset was divided into a training set, a validation set, and a test set, with proportions of 60%, 20%, and 20%, respectively.

[0121] Step 3.2, Building the Proxy Model. After preparing the training dataset in Step 3.1, we proceed to the proxy model building phase. The following are the detailed steps for model building:

[0122] Proxy model architecture design:

[0123] Input layer design: The input layer of the data-driven model receives multiple long-sequence features, including long-sequence hydrological and water quality data input to the three-dimensional hydrodynamic-water quality model of the large reservoir, as well as multi-layer water quality concentration data output by the three-dimensional hydrodynamic-water quality model of the large reservoir; if the dimension of the input features is n, then the input layer contains n neurons.

[0124] Hidden layer design: To capture the complex nonlinear relationship between water quality and environmental drivers, the model contains multiple hidden layers and uses ReLU as the activation function. ReLU can effectively alleviate the vanishing gradient problem and improve the convergence speed.

[0125] Output layer design: The model's output layer consists of multiple neurons used to predict water quality-related indicators. Single-point prediction is employed, meaning the output shows the water quality change over time at a representative location / layer. Since water quality prediction is a regression problem, the output layer uses a linear activation function, suitable for numerical prediction.

[0126] Training process:

[0127] Training dataset: Using the training set from step 3.1, the input features are meteorological data, hydrological data, and pollution load data, and the target variable is the water quality change process (such as surface water quality, vertical water quality, etc.). The loss function is the mean squared error (MSE), which calculates the difference between the predicted values ​​of the surrogate model and the simulated output values ​​of the hydrodynamic-water quality model.

[0128] Model performance optimization: The Adam optimizer is employed. This optimization algorithm combines adaptive learning rate and momentum methods to improve the efficiency and convergence speed of surrogate model training. The Adam algorithm updates the parameters of each layer according to the following update rules:

[0129] Regularization: To avoid overfitting, regularization techniques, also known as weight decay, are used. This involves adding a regularization term to the loss function to control model complexity.

[0130] Training strategy: An early stopping strategy is used to prevent overfitting. Training is stopped early when performance on the validation set no longer improves. This strategy effectively saves training time while avoiding overfitting to the training data.

[0131] Step 3.3, Model Validation and Evaluation:

[0132] To ensure the accuracy and generalization ability of the model, the surrogate model underwent various validations and evaluations. Cross-validation was primarily used to ensure the stability and reliability of the surrogate model. Specifically:

[0133] Cross-validation: This method uses K-fold cross-validation to evaluate the model. The data is divided into K subsets, and one subset is used as the validation set each time, while the remaining subsets are used for training. Finally, the average performance of the model is evaluated. This method effectively reduces the random errors introduced by data partitioning.

[0134] Evaluation metrics: The main metrics used are root mean square error (RMSE) and Nash efficiency coefficient (NSE). The closer the value is to 1, the better the model's performance.

[0135] Performance reference of the proxy model Figure 3 The horizontal axis represents the predicted values ​​from the surrogate model, and the vertical axis represents the simulated values ​​from the hydrodynamic-water quality model of the Shenda Reservoir. All data points are closely distributed on both sides of the 1:1 line, indicating that the surrogate model has high prediction accuracy and reliability and can accurately reflect the changing patterns of the actual physical processes.

[0136] Step 4: Construct a reservoir optimal scheduling model and couple the data-driven surrogate model obtained in Step 3 with the reservoir optimal scheduling as a water quality response prediction tool. This model satisfies multiple water quality objectives while considering power generation needs, and proposes a suitable scheduling process for decision-makers. Specifically:

[0137] Step 4.1, Define decision variables:

[0138] Determining Decision Variables: To ensure that predetermined ecologically relevant water quality and power generation targets can be achieved under different inflow scenarios, the reservoir scheduling process needs to be treated as a decision variable, making it applicable under uncertainty. The scheduling rules consist of a combination of target reservoir capacity, output coefficient, and discharge location set for each scheduling period, used to determine the correspondence between unit output and reservoir capacity.

[0139] Step 4.2, Set optimization goals:

[0140] Determine the power generation target: Maximize the average annual power generation of the Shenda Reservoir as the power generation benefit optimization target, and provide an economic evaluation basis for subsequent dispatching schemes, as shown in formula (4).

[0141] Determine the downstream water quality targets: Considering that the main release of dissolved nitrogen and phosphorus in the reservoir area during the non-flood season, while the main retention and interception of particulate nitrogen and phosphorus during the flood season, different water quality targets are set for different time periods to alleviate the problems of eutrophication in the reservoir area and oligotrophication downstream. During the non-flood season, the control target is mainly the concentration of dissolved nitrogen and phosphorus in the reservoir water, controlling their concentration as low as possible to improve the self-purification capacity of the water body and reduce the risk of eutrophication in the reservoir area; during the flood season, the control target is mainly the downstream flux of biogenic nitrogen and phosphorus, optimizing the discharge strategy to achieve the transport of as much effective biogenic material as possible downstream, thereby alleviating the oligotrophication phenomenon in the downstream water body. See formulas (5)-(8).

[0142] Step 4.3, Set constraints:

[0143] Constraints: Optimal reservoir scheduling generally follows constraints such as water balance, reservoir capacity, reservoir discharge, power generation flow, reservoir output, and scheduling rule parameter range. See formulas (9)-(12).

[0144] Step 4.4, Define the solution algorithm:

[0145] Algorithm Selection for Optimization: Considering the multi-objective nature of the reservoir optimization scheduling model, the heuristic NSGA-III algorithm, which balances computational efficiency and solution quality, was chosen to solve the optimization scheduling model. Key algorithm parameters were set as follows: population size of 200, maximum number of generations of evolution of 500, crossover probability of 0.9, mutation probability of 1 / D (where D is the dimension of the decision variable), reference point generation using the Das and Dennis normalized boundary crossover method, crossover distribution index set to 30, mutation distribution index set to 20, and simulated binary crossover and multinomial mutation operators employed. This parameter combination effectively improves the convergence performance and distribution uniformity of the algorithm in high-dimensional objective spaces while ensuring population diversity.

[0146] By solving the multi-objective optimization scheduling model, a set of scheduling rules that meet the power generation demand and water quality requirements were obtained, providing a scientific basis and technical support for the comprehensive management of water resources and water environment to alleviate the retention of biological materials in the reservoir area and improve the water quality of downstream areas.

[0147] An ecological scheduling method for alleviating biomass retention in the Shenzhen-Dalian Reservoir is implemented based on the Shenzhen-Dalian Reservoir ecological scheduling system. This system includes a data acquisition module, a Shenzhen-Dalian Reservoir hydrodynamic-water quality model construction module, a data-driven proxy model construction module, and a reservoir scheduling optimization module. Specifically:

[0148] Data acquisition module: used to acquire characteristic parameters of the deep and large reservoirs in the research watershed and various types of hydrological and water quality monitoring data, providing basic input for hydrodynamic and water quality simulation of the deep and large reservoirs.

[0149] The three-dimensional hydrodynamic-water quality model construction module of Shenzhen Reservoir: Constructs a three-dimensional hydrodynamic module and a water quality module for the Shenzhen Reservoir to study the reservoir. By solving the continuity equation, momentum equation and water quality component transport equation, it simulates the spatiotemporal distribution characteristics of water quality indicators of the reservoir under historical operating conditions, and couples the two to form a reliable three-dimensional hydrodynamic-water quality model of Shenzhen Reservoir.

[0150] Data-driven surrogate model building module: Based on the simulation results of the three-dimensional hydrodynamic-water quality model of Shenzhen Reservoir and related input features and output results, a surrogate model suitable for water quality prediction is built;

[0151] Reservoir scheduling optimization module: Coupled with a data-driven proxy model, a reservoir optimization scheduling model is constructed, and a set of scheduling rules is derived under the premise of meeting power generation demand and water quality requirements.

[0152] The above embodiments are merely illustrative of the implementation methods of the present invention, but should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the protection scope of the present invention.

Claims

1. An ecological regulation method for alleviating the retention of biogenic materials in deep and large reservoirs, characterized in that, The ecological scheduling method includes the following steps: The first step is to collect characteristic parameters of deep and large reservoirs in the research basin and various types of hydrological and water quality monitoring data; The second step is to construct a three-dimensional hydrodynamic-water quality model of the deep reservoir. Based on the characteristic parameters of the Shenzhen Reservoir and various hydrological and water quality monitoring data collected in the first step, a three-dimensional hydrodynamic module and a water quality module of the Shenzhen Reservoir are constructed, and the coupling coefficient between the two is determined to form a three-dimensional hydrodynamic-water quality model of the Shenzhen Reservoir. Step 2.1: Based on the characteristic parameters of the deep reservoir and the collection of various hydrological and water quality monitoring data in the first step, the planar and vertical grids are divided, and the topographic and water depth data are input and the boundary conditions are set, thereby constructing the three-dimensional hydrodynamic module of the deep reservoir. Step 2.2: Selecting water quality-related processes and setting key parameters to construct the water quality module; Step 2.3: Couple the water quality module with the three-dimensional hydrodynamic module of the Shenzhen Reservoir to form a three-dimensional hydrodynamic-water quality model of the Shenzhen Reservoir, as detailed below: The three-dimensional hydrodynamic module of the deep reservoir obtained in step 2.1 and the water quality module obtained in step 2.2 are coupled together by a loose coupling mechanism to achieve dynamic simulation of material migration and change in complex water bodies; First, the three-dimensional hydrodynamic module of Shenzhen Reservoir simulates and outputs key hydrodynamic parameters such as flow velocity, water depth, temperature, and turbulent diffusion coefficient. Then, the key hydrodynamic parameters are passed to the water quality module as forced input in each spatial grid and at each time step, driving it to dynamically simulate the convection, diffusion, and reaction processes of matter. Thus, the construction of the three-dimensional hydrodynamic-water quality model of Shenzhen Reservoir is completed. Step 2.4: Based on the characteristic parameters of the Shenda Reservoir and various types of hydrological and water quality monitoring data collected in the first step, the three-dimensional hydrodynamic-water quality model of the Shenda Reservoir is calibrated and verified in an integrated manner, encompassing multiple processes, multiple elements, multiple time and space, and multiple indicators. Step 2.5: Based on the calibrated and verified three-dimensional hydrodynamic-water quality model of the Shenzhen Reservoir, organize its input and simulation output results to form a training dataset for building the subsequent data-driven surrogate model; The third step is to build a data-driven proxy model based on the three-dimensional hydrodynamic-water quality model of the deep reservoir constructed in the second step. Step 4: Construct a reservoir optimization scheduling model, and couple the data-driven surrogate model obtained in Step 3 with the reservoir optimization scheduling model as a water quality response prediction tool, and solve it to obtain scheduling rules that meet power generation needs and water quality requirements.

2. The ecological regulation method for alleviating the retention of biogenic materials in deep and large reservoirs according to claim 1, characterized in that, In the first step, the characteristic parameters of the deep reservoir include the digital elevation model (DEM), dam height, total reservoir capacity, normal water level, and water level-capacity curve. The various types of hydrological and water quality monitoring data include: meteorological data, hydrological data, pollution load data, inflow data, outflow data, water level data, surface water temperature data, vertical water temperature data, surface water quality data, vertical water quality profile data, and point source and non-point source pollution data.

3. The ecological regulation method for alleviating the retention of biogenic materials in deep and large reservoirs according to claim 2, characterized in that, In the second step: Step 2.1 specifically involves: Mesh type selection: Use curvilinear coordinates for planar meshing; select the σ coordinate system for vertical meshing; Topographic and water depth data input: Based on the spatial data of the digital elevation model (DEM), extract the elevation data of the entire reservoir area and assign accurate elevation values ​​to each grid cell in the reservoir area; Boundary condition settings: mainly include input data, initial background value settings, and warm-up period selection; This completes the construction of the three-dimensional hydrodynamic module for the Shenzhen Reservoir. Step 2.2 specifically involves: Selection of water quality-related processes: comprehensively considering the migration and transformation processes of key water quality components such as dissolved oxygen, ammonia nitrogen, nitrate nitrogen, orthophosphate, particulate nitrogen, and particulate phosphorus in water bodies; Key parameter settings: gravitational acceleration, water density, and air density; corresponding drag coefficients and wind speed thresholds are set for different wind speed ranges; parameters related to friction coefficient, horizontal eddy viscosity coefficient, and eddy diffusion coefficient are also included; in terms of heat exchange at the water-air interface, Secchi depth, Dalton number, and Stanton number are key parameters controlling the evaporation and heat exchange process at the water surface; in terms of water quality parameter settings, nitrogen and phosphorus decomposition rate constants and sedimentation velocity parameters are set to simulate the conversion process of particulate nitrogen and phosphorus to dissolved nitrogen and phosphorus. This completes the construction of the water quality module.

4. The ecological regulation method for alleviating the retention of biogenic materials in deep and large reservoirs according to claim 3, characterized in that, Specifically, step 2.4 involves: In terms of multiple processes: the focus is on the dynamic linkage and physical consistency of key processes in hydrodynamic evolution, water temperature structure, and water quality transformation; In terms of multiple factors: water level, flow rate, water temperature, total nitrogen, total phosphorus, dissolved oxygen, and biochemical oxygen demand are included as variables; In terms of multiple time and space: it covers high-frequency water level and water temperature data on a daily scale throughout the year; In terms of multiple indicators: the Nash efficiency coefficient (NSE), coefficient of determination (R²), root mean square error (RMSE), and absolute relative deviation (ADR%) are used to comprehensively evaluate the performance and physical consistency of the three-dimensional hydrodynamic-water quality model of the Shenda Reservoir under different processes, elements, and spatiotemporal scales.

5. The ecological regulation method for alleviating the retention of biogenic materials in deep and large reservoirs according to claim 4, characterized in that, The third step is specifically as follows: Step 3.1, Organize and divide the training dataset: Based on the training dataset of the data-driven agent model obtained in step 2.5, it is divided into training set, validation set and test set; Step 3.2, Construction of the proxy model: Step 3.2.1, Proxy Model Architecture Design: Input layer design: The input layer of the surrogate model receives multiple long sequence features, including hydrological and water quality data input to the three-dimensional hydrodynamic-water quality model of Shenzhen Reservoir, as well as multi-layer water quality concentration data output by the three-dimensional hydrodynamic-water quality model of Shenzhen Reservoir; If the input feature dimension is n, then the input layer contains n neurons; Hidden layer design: It contains multiple hidden layers and uses ReLU as the activation function; Output layer design: The output layer consists of multiple neurons and is used to predict water quality indicators; Step 3.2.2, Training Process: Training dataset: Using the training set from step 3.1, input the meteorological data, hydrological data, and pollution load data collected in step 1 as features, with the target variable being the water quality change process; The loss function uses mean squared error (MSE) to calculate the difference between the predicted values ​​of the surrogate model and the output values ​​of the three-dimensional hydrodynamic-water quality model of the Shenzhen Reservoir: (1) in, For the first The predicted value for each sample, These are actual observations. The total number of samples, Indicates the sample index; Proxy model performance optimization: Using the Adam optimizer, the parameters of each layer are updated according to the following update rules: (2) in, For learning rate, and For estimates of the first and second moments, To prevent division by zero of constants; For the first Model parameters obtained through time-period iteration; For the first Model parameters obtained over a time period; Regularization: Regularization techniques are used to control complexity by adding a regularization term to the loss function. (3) in, The regularization coefficient is . These are the weight parameters in the model; Total loss; Training strategy: Employ an early stopping strategy to prevent overfitting; Step 3.3: Use cross-validation to ensure the stability and reliability of the proxy model.

6. The ecological regulation method for alleviating the retention of biogenic materials in deep and large reservoirs according to claim 5, characterized in that, Step 3.3 specifically involves: evaluating the surrogate model using the K-fold cross-validation method, dividing the data into K subsets, using one subset as the validation set each time, and using the remaining subset for training, and finally evaluating the average performance of the surrogate model; the evaluation metrics include root mean square error (RMSE) and Nash efficiency coefficient (NSE), with values ​​closer to 1 indicating better performance.

7. The ecological regulation method for alleviating the retention of biogenic materials in deep and large reservoirs according to claim 6, characterized in that, The fourth step is specifically as follows: Step 4.1, Define decision variables: The reservoir scheduling process is used as a decision variable. The scheduling rules are composed of the target reservoir capacity, output coefficient and discharge position set for each scheduling period, which are used to determine the correspondence between unit output and reservoir capacity. Step 4.2, Set optimization goals: Determine the power generation target: Maximize the multi-year average power generation of the Shenda Reservoir as the power generation efficiency optimization target; (4) in, This represents the total number of scheduling periods. Indicates the calculation period; To contribute power to the power plant; This represents the average annual power generation over many years. For time step; Determine the water quality targets for outflow: set different water quality targets for different time periods; during the non-flood season, the control target is mainly the concentration of dissolved nitrogen and phosphorus in the reservoir water; during the flood season, the control target is mainly the outflow flux of nitrogen and phosphorus from biological sources. (5) (6) (7) (8) in, For the reservoir in The outflow rate at any given moment; The layer where the reservoir discharges water is located. Time period TP concentration within; The layer where the reservoir discharges water is located. Time period TN concentration within; The average TN concentration across multiple layers in the reservoir area; CS TP The average TP concentration across multiple layers in the reservoir area; N represents the total number of monitoring times, and n represents the index of the number of monitoring times. The TP flux through the reservoir discharge section; The TN flux through the reservoir discharge section; Step 4.3: Set constraints, including water balance constraints, reservoir capacity constraints, reservoir discharge constraints, power generation flow constraints, reservoir output constraints, and scheduling rule parameter range constraints. Step 4.4, Define the solution algorithm: Optimization algorithm selection: Considering that the reservoir optimization scheduling model is multi-objective, the heuristic algorithm NSGA-III is selected to solve it; By solving a multi-objective reservoir optimization scheduling model, a set of scheduling rules that meet both power generation demand and water quality requirements are obtained.

8. The ecological regulation method for alleviating the retention of biogenic materials in deep and large reservoirs according to claim 7, characterized in that, The ecological scheduling method described above is implemented based on the Shenzhen-Daya Reservoir ecological scheduling system, which includes a data acquisition module, a Shenzhen-Daya Reservoir hydrodynamic-water quality model construction module, a data-driven proxy model construction module, and a reservoir scheduling optimization module.

9. An ecological regulation method for alleviating the retention of biogenic materials in deep and large reservoirs according to claim 8, characterized in that, The aforementioned ecological scheduling system for the deep reservoir: Data acquisition module: used to acquire characteristic parameters of deep and large reservoirs in the research watershed and various types of hydrological and water quality monitoring data; The three-dimensional hydrodynamic-water quality model construction module of Shenzhen Reservoir: Constructs a three-dimensional hydrodynamic module and a water quality module for the Shenzhen Reservoir to study the reservoir. By solving the continuity equation, momentum equation and water quality component transport equation, it simulates the spatiotemporal distribution characteristics of water quality indicators of the reservoir under historical operating conditions, and couples the two to form a reliable three-dimensional hydrodynamic-water quality model of Shenzhen Reservoir. Data-driven surrogate model building module: Based on the simulation results of the three-dimensional hydrodynamic-water quality model of Shenzhen Reservoir and related input features and output results, a surrogate model suitable for water quality prediction is built; Reservoir scheduling optimization module: Coupled with a data-driven proxy model, a reservoir optimization scheduling model is constructed, and a set of scheduling rules is derived under the premise of meeting power generation demand and water quality requirements.

Images