An ecological regulation method for river-lake systems coupling hydrological responses of river basins

Abstract

An ecological operation method for river-lake systems coupling basin hydrological responses, belonging to the field of reservoir ecological operation. The steps are as follows: First, select representative hydrological stations in the study area and draw a topological map of basin hydrological evolution; Second, gradually construct a basin hydrological response model to simulate the response process of lake hydrological regime to the discharge of upstream reservoirs; Third, propose ecological evaluation indicators to quantify the hydrological-ecological response of lakes; Fourth, construct a multi-objective optimal operation model for river-lake systems; Fifth, optimize the multi-objective optimal operation model for river-lake systems; Finally, based on the fuzzy optimization method, recommend the optimal reservoir operation rules that meet the decision-making preferences. The present invention can, while ensuring the beneficial effects of reservoir power generation, water supply, etc., minimize the adverse ecological impacts of concentrated water storage in upstream reservoirs on downstream lakes, recommend reservoir operation rules with maximized comprehensive benefits, and provide a scientific operation basis and decision-making support for the green development of river-lake systems.

Description

Technical Field

[0001] The present invention belongs to the field of reservoir ecological operation, and relates to an ecological operation method for river-lake systems that couples basin hydrological responses. Background Art

[0002] Lakes are an important part of the global freshwater resources and play important roles in runoff regulation, flood control safety, agricultural irrigation, regional ecology, etc. However, in recent years, many water conservancy projects have been planned and built in major river basins to improve the water resource utilization efficiency. A large amount of concentrated water storage has significantly reduced the downstream discharge during the reservoir water storage period, resulting in a decline in lake water levels, an earlier onset of the dry season, frequent drought events, and further deterioration of the lake ecological environment, such as a reduction in biodiversity and a weakening of the ecological regulation function.

[0003] The optimization of reservoir operation plans is crucial for improving the adverse impacts on lake ecology and promoting the sustainable development of river-lake systems. In recent years, scholars have increasingly incorporated ecological water requirements into the operation objectives and recommended optimized operation processes or operation rules for reservoir operation management. For example, He Zhongzheng et al. (An optimized reservoir power generation operation method considering the protection of the spawning of four major Chinese carps [P]. Jiangxi Province: CN202211410932.4, 2023-04-14.) disclosed an optimized reservoir power generation operation method considering the protection of the spawning of four major Chinese carps. Taking the continuous flow increase rate and continuous days required for the protection of the spawning of four major Chinese carps as the downstream discharge constraints of the reservoir, an optimized reservoir power generation operation model was constructed to optimize the optimal reservoir power generation operation process considering the protection of the spawning of four major Chinese carps. Li Xinan et al. (An optimization method for a water supply-power generation-ecology multi-objective operation chart based on ecological flow [P]. Hubei Province: CN201911080419.1, 2020-02-28.) disclosed an optimization method for a water supply-power generation-ecology multi-objective operation chart based on ecological flow. Taking the shortage and overflow of the ecological flow threshold of the downstream river channel by the reservoir downstream discharge as the ecological operation objective, the water supply operation rules of the reservoir were optimized to coordinate multiple objectives such as the water supply, ecology, and power generation of the reservoir.

[0004] However, most of the existing technologies focus on the ecological needs in the reservoir area or the downstream river channel, and the satisfaction degree of its ecological objectives can be directly derived from the water level or flow rate of the reservoir. However, the hydrological evolution between the reservoir and the lake is relatively complex and is affected by multiple factors such as river channel topography and water volume exchange between the river and the lake. The ecological objectives of the lake cannot be directly calculated from the reservoir discharge process. This means that only based on the traditional optimization model, it is impossible to meet the scheduling requirements that simultaneously consider the ecological objectives of the lake and the beneficial objectives of the reservoir. It is necessary to further construct and couple a basin hydrological response model on this basis to simulate the complex hydraulic relationship of the river-lake system, so as to realize the calculation and optimization of ecological objectives. The coupling of the basin hydrological response model significantly increases the complexity and modeling difficulty of the multi-objective optimization scheduling model, and puts forward higher requirements for the multi-objective optimization scheduling of the river-lake system and the analysis of the high-dimensional non-linear competition and cooperation relationship between objectives. Summary of the Invention

[0005] Aiming at the problems existing in the prior art, the present invention provides an ecological scheduling method for a river-lake system coupled with a basin hydrological response, which while ensuring the beneficial effects of the reservoir such as power generation and water supply, maximally weakens the adverse ecological impact of the concentrated water storage of the upstream reservoir on the downstream lake, recommends the reservoir scheduling rules with the maximized comprehensive benefits, and provides a scientific scheduling basis and decision-making support for the green development of the river-lake system.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] An ecological scheduling method for a river-lake system coupled with a basin hydrological response, comprising the following steps:

[0008] Step 1, select representative hydrological stations in the study area and draw a topological map of the basin hydrological evolution;

[0009] Step 2, gradually construct a basin hydrological response model to simulate the response process of the lake hydrological situation to the upstream reservoir discharge;

[0010] Step 3, propose ecological evaluation indicators to quantify the lake hydrological-ecological response;

[0011] Step 4, construct a multi-objective optimization scheduling model for the river-lake system;

[0012] Step 5, optimize the multi-objective optimization scheduling model for the river-lake system;

[0013] Step 6, based on the fuzzy optimization method, recommend the optimal reservoir scheduling rules that meet the decision-making preferences.

[0014] Further, the specific implementation method of Step 1 is as follows:

[0015] Step 1.1, Collect basin information and analyze characteristics. Based on the topographic map and river system map of the basin, collect the geographical information of the basin, clarify the distribution of the basin river system, the connection relationship of the river channels, and the water flow direction and path, and analyze the river confluence and diversion characteristics, the water volume exchange characteristics between rivers and lakes, the inflow and outflow characteristics of the lakes, and the seasonal change characteristics of the lake morphology in the study area.

[0016] Step 1.2, Select representative hydrological stations in the basin. Based on the existing hydrological station network in the basin, select hydrological stations with long-term, continuous, and consistent historical measured data as the representative stations for subsequent construction of the basin hydrological response model. The selection of representative stations should follow the following principles:

[0017] Step 1.2.1, Reservoir representative station: A reservoir outflow station should be set at the dam site of the reservoir;

[0018] Step 1.2.2, River channel representative station: At least one representative station should be selected for each main and tributary river channel. Specifically, at the section where there is a tributary confluence or diversion, representative stations should be set both upstream and downstream of the confluence point; in the section where there is no tributary confluence or diversion, generally one representative station can be set, and in cases where the river channel topography is complex, a small number of representative stations can be added appropriately, but the geographical positions of the stations should not be too close to each other;

[0019] Step 1.2.3, Lake representative station: At least one representative station should be selected inside the lake area, and the specific number is determined according to the spatial hydrological response characteristics inside the lake. If there is no obvious difference in the spatial hydrological response inside the lake, that is, the water level change processes at different spatial positions of the lake are almost the same, then one lake representative station can be selected; if there are obvious differences in the spatial hydrological response inside the lake, that is, the water level heights and response times at different spatial positions of the lake are inconsistent, then the lake is divided into multiple lake areas according to its topographic characteristics and hydrological response characteristics, and one representative station is set in each lake area.

[0020] Step 1.3, Draw the topological map of basin hydrological evolution. Generalize the basin river system structure, use the representative stations selected in Step 1.2 as nodes, use straight lines to connect to represent the hydraulic connection between nodes, and use arrows to indicate the flow direction; use rectangles to mark the lake boundaries, and indicate the names of the reservoirs and each representative station in the study area to form a topological map of basin hydrological evolution that can accurately reflect the spatio-temporal correlation between reservoirs, rivers, and lakes.

[0021] Furthermore, the specific implementation method of Step 2 is as follows:

[0022] Step 2.1, Decompose the study area and set up hydrological response sub-models step by step. Based on the topological map of basin hydrological evolution drawn in Step 1.3, decompose the study area, set up hydrological response sub-models for each group of adjacent representative stations respectively, and quantify the basin hydrological response step by step.

[0023] Step 2.1.1: For the reservoir representative station and the river channel representative station, the hydrological response sub-model uses the hydrological characteristics of the upstream representative station adjacent to the representative station as input variables and the hydrological characteristics of the downstream representative station as output variables. Through two steps of determining the function type and optimizing the parameter values, the hydrological response sub-model is constructed in sequence. The hydrological characteristics of each representative station can be the water level, water depth, flow rate, or water volume characteristics of the representative station.

[0024] Step 2.1.2: For the lake representative station, when there is only one lake representative station in the study area, the hydrological characteristics of the river channel representative station where the lake inflow is located are used as input variables, and the hydrological characteristics of the lake representative station are used as output variables to construct the hydrological response sub-model of the lake, quantifying the response of the lake hydrological regime to the inflowing runoff. When there are multiple lake representative stations in the study area, first select one of the lake representative stations and construct a hydrological response sub-model with the hydrological characteristics of the river channel representative station where the lake inflow is located; then, using the hydrological characteristics of the selected lake representative station as input and the hydrological characteristics of other lake representative stations as output, construct a hydrological response sub-model to quantify the hydrological correlation relationships among multiple lake areas within the lake.

[0025] Step 2.2: Determine the function type of the hydrological response sub-model.

[0026] Step 2.2.1: The candidate function types include but are not limited to linear functions, polynomial functions, multivariate polynomial functions, or piecewise functions.

[0027] Step 2.2.2: Use the k-fold cross-validation method to train and test all candidate function types. Using the historical measured data of the hydrological characteristics of the selected representative station as the data set, evenly divide it into k subsets. Each time, select one subset as the test set, and the remaining k - 1 subsets as the training set, and perform training and testing on each candidate function type.

[0028] Step 2.2.3: Conduct accuracy evaluation and optimization for the candidate function types. Select the Pearson correlation coefficient CC, root mean square error RMSE, and relative bias RB as statistical indicators, and quantify the simulation accuracy of each candidate function type with the average value of its k training and testing times. Select the function type with the optimal simulation accuracy for the hydrological response sub-model.

[0029] Step 2.3: Optimize the parameter values of the hydrological response sub-model. Use all the data sets in Step 2.2.2 to train the function type selected in Step 2.2.3, and optimize and determine the parameter values of the hydrological response sub-model.

[0030] Step 2.4: Integrate the hydrological response sub-models to form a complete basin hydrological response model.

[0031] Based on the function type determined in Step 2.2 and the optimized parameter values in Step 2.3, determine the structure of the hydrological response sub-model, and integrate the hierarchically constructed hydrological response sub-models into a complete basin hydrological response model.

[0032] When applying the basin hydrological response model, use the hydrological response sub-models step by step in the order from upstream to downstream. The output of the upstream sub-model is used as the input data for the downstream sub-model to simulate the lake water level at each time period. For the discharge of the upstream reservoir and the inflow from the intermediate area the step-by-step dynamic response of which can be characterized as:

[0033] (1)

[0034] Furthermore, the specific implementation method of Step 3 is as follows:

[0035] The basin hydrological response model can simulate the dynamic response process of the lake water level. On this basis, Step 3 quantifies the ecological benefits of the lake at different water levels from two aspects: the degree of ecological satisfaction and the degree of ecological water shortage, based on the ecological water demand process of the lake, that is, the hydrological-ecological response of the lake.

[0036] Step 3.1, quantify the degree of ecological satisfaction of the lake.

[0037] Set the ecological satisfaction index of the lake to measure whether the water levels in each lake area of the lake can meet the ecological water level demand at each time period; then, based on the importance weight of the lake area, weight the ecological satisfaction of each lake area to obtain the ecological guarantee degree of the lake at the time period. The formula is as follows:

[0038] (2)

[0039] (3)

[0040] Among them, represents the degree of ecological satisfaction of the lake at time t, which is the weighted average of the ecological satisfaction of each lake area. represents the ecological satisfaction of the i-th lake area at time t, and the value is 0 or 1. When the simulated water level LV i,t of the representative station in the i-th lake area during the time period is not lower than the ecological demand D i,t it is considered that the lake area meets the ecological demand, and W i,t is taken as 1, otherwise 0. represents the importance weight of lake area i, which is determined based on expert experience, and N zone represents the number of partitions included in the lake, satisfying and the constraints of.

[0041] Step 3.2, Quantify the degree of ecological water shortage in lakes.

[0042] Take the square of the ratio of the ecological water level to the simulated water level in the lake area as the index of the degree of ecological water shortage, and measure the degree of ecological water shortage in each lake area during the period. If there is no ecological water shortage in the lake area, the index value is in the interval (0, 1]; otherwise, if there is ecological water shortage in the lake area, the index is greater than 1, and the closer the value is to positive infinity, the lower the simulated water level during the period and the more serious the water shortage. The formula is as follows:

[0043] (4)

[0044] (5)

[0045] Among them, represents the degree of ecological water shortage of the lake in the t period, which is the weighted average of the degrees of ecological water shortage in each lake area. represents the degree of ecological water shortage of the i-th lake area in the t period, and the value range is (0, +∞).

[0046] Furthermore, the specific implementation method of Step 4 is as follows:

[0047] Step 4.1, Set the reservoir optimal operation chart and decision variables.

[0048] Based on the conventional reservoir operation rules, set the reservoir optimal operation chart in multi-objective optimal operation, in the form as Figure 2 shown. The reservoir optimal operation chart includes three operation lines, namely the reduced output line, the guaranteed output line, and the increased output line. The three output lines divide the operation interval into four output areas, namely the reduced output area, the guaranteed output area, the increased output area 1, and the increased output area 2. The output coefficients of each output area are , , and , satisfying the constraint . When conducting reservoir power generation operation based on the operation chart, the output in each period is determined by the output area where the water level at the beginning of the period is located, and the output size is equal to the product of the output coefficient and the guaranteed output.

[0049] Therefore, the reservoir optimal operation chart includes two types of decision variables (optimized and determined through Step 5), namely the water level variables of each operation period that determine the position of the output line, and the output coefficient variables that determine the output size of each period.

[0050] Step 4.2, Set the objective function. The operation objectives include the lake ecological objective and the reservoir beneficial operation objective. Among them, the lake ecological objective is the maximization objective of the lake ecological satisfaction degree and the minimization objective of the lake ecological water shortage degree, and the reservoir beneficial operation objective is the maximization objective of the dimensionless reservoir power generation and the maximization objective of the end-of-flood season water storage. The definitions and formulas of each objective are as follows:

[0051] Step 4.2.1, maximize the goal of lake ecological satisfaction. To achieve the maximization of lake ecological satisfaction, the lake ecological satisfaction levels in each time period in Step 3.1 are weighted by time period. The time period importance weights are allocated based on the current ecological situation of the basin and the basin operation requirements. Specifically, for critical time periods with more frequent and severe ecological water shortages and greater attention in operation requirements, higher weights should be assigned; while for time periods that can better meet ecological needs in the current situation, the importance weights should be appropriately reduced or even set to 0. The formula is as follows:

[0052] (6)

[0053] Where, is the weighted lake ecological satisfaction level for each operation time period, and are the overall lake ecological satisfaction level and the ecological satisfaction of the i-th lake area at time t, respectively. is the time period importance weight; N period is the number of operation time periods, satisfying and . is the importance weight of lake area i, N zone is the number of lake partitions, satisfying and .

[0054] Step 4.2.2, minimize the goal of lake ecological water shortage degree: Weight the lake ecological water shortage degree indicators in each time period in Step 3.2 by time period.

[0055] (7)

[0056] Where, is the weighted lake ecological water shortage degree for each operation time period, and are the overall ecological water shortage degree of the lake and the ecological water shortage degree of the i-th lake area at time t, respectively.

[0057] Step 4.2.3, maximize the goal of reservoir power generation: Maximize the annual average power generation over the years, and dimensionlessize the goal based on the initial design or the annual average power generation statistically obtained from historical data.

[0058] (8)

[0059] Where, is the dimensionless annual average power generation of the reservoir over the years; N output is the guaranteed output of the reservoir, is the duration of the operation time period, is the output coefficient for a time period, which is determined according to the output area where the water level is at the beginning of the time period, and takes values of , , or ; is the annual average power generation, and N year is the number of dispatching years.

[0060] Step 4.2.4, maximizing the reservoir water storage at the end of the flood season: The reservoir has the maximum average annual water storage at the end of the flood season, which is characterized by the average annual storage capacity at the end of the reservoir's water storage period.

[0061] (9)

[0062] Among them, is the dimensionless average annual water storage at the end of the flood season; V n is the storage capacity at the end of the water storage period of the nth year, and V max is the total storage capacity of the reservoir.

[0063] Step 4.3, setting constraint conditions. The constraints of the multi-objective optimal operation model include: water balance constraint, reservoir water level constraint, reservoir discharge flow constraint, hydropower station output constraint, and variable non-negativity constraint.

[0064] Furthermore, the specific implementation method of Step 5 is as follows:

[0065] The optimization and solution of the multi-objective optimal operation model follow the overall idea of "simulating the dispatching process - optimizing the dispatching rules" and are carried out based on the optimization algorithm.

[0066] Step 5.1, simulating the dispatching process. Based on the values of the decision variables, determine the current reservoir optimal operation chart. Taking the inflow runoff process as the input, conduct water balance calculations to derive the reservoir water level, discharge flow, and output process. Input the reservoir discharge process into the basin hydrological response model constructed in Step 2 to derive the spatial water level response process of the downstream lake, and then evaluate the lake ecological level based on the indicators proposed in Step 3, thereby deriving the values of the lake ecological target and the reservoir beneficial operation target under the current operation chart, and returning them to the optimization stage.

[0067] Step 5.2, optimizing the dispatching rules. Based on the selected optimization algorithm, generate an initial solution set; evaluate and screen the solution set based on the objective values calculated in the dispatching process simulation stage of Step 5.1; through iterative optimization, gradually improve the solution set until the number of iterations is equal to the preset threshold or the variation range of the objective value is less than the preset threshold; finally, output the optimal approximate Pareto solution set, as well as the corresponding reservoir dispatching process and lake response process. The approximate Pareto solution set includes one or more feasible solutions, and each feasible solution corresponds to a set of values of the decision variables described in Step 4.1, and the reservoir optimal operation chart corresponding to the current feasible solution can be drawn based on it.

[0068] Further, the specific implementation of step 6 is as follows:

[0069] Step 6.1: Calculate the improvement ability of the optimized scheduling graph compared with the conventional scheduling. Statistically calculate the optimal values and average values of the scheduling objectives and important scheduling features (such as lake water levels, reservoir power generation, etc.) of all feasible solutions in the approximate Pareto solution set. Calculate the relative percentage of improvement for each scheduling objective compared with the conventional reservoir scheduling.

[0070] Step 6.2: Calculate the relative membership degree of the scheduling objectives of the Pareto solution set. Since the value ranges of multiple scheduling objectives are different, standardize them separately and use formula (10) to obtain the relative membership degree of the scheduling objectives of the Pareto solution set.

[0071] (10)

[0072] Where, and are respectively the objective value and relative membership degree of the i-th scheduling objective of the j-th feasible solution, and are respectively the minimum value and maximum value of the i-th scheduling objective.

[0073] Step 6.3: Calculate the weights of the scheduling objectives according to the decision-making preference. According to the decision-making preference, compare the importance between two objectives pairwise, use "equally important, slightly important, somewhat important, more important, significantly important, highly important, very important, extremely important, supremely important, incomparable" as fuzzy tone operators to determine the fuzzy scale values and describe the relative importance degree between objectives. Among them, the fuzzy scale values of equally important and incomparable are 0.5 and 1.0 respectively, and the fuzzy scale values of other fuzzy tones are calculated by linear interpolation method. Substitute the given fuzzy tone operators into formula (11) to obtain the importance degrees of different objectives, and after normalizing them, obtain the weights of each scheduling objective .

[0074] (11)

[0075] Where, represents the importance degree of all scheduling objectives; , , and respectively represent the importance degrees of objective 1 to objective 4, where objective 1 is the most important objective considered by the decision-maker; , and respectively represent the fuzzy scale values of the 2nd, 3rd, and 4th objectives compared with objective 1, which are determined by the fuzzy tone operators determined by the decision-maker.

[0076] Step 6.4, calculate the relative membership degree, and recommend the optimal scheduling diagram that meets the decision-making preference. Assume that the relative membership degrees of the optimal solution and the worst solution are and respectively. Then the relative membership degree of the j-th feasible solution with respect to the optimal solution is:

[0077] (12)

[0078] where represents the relative membership degree of the j-th feasible solution; represents the weight of the k-th scheduling objective calculated in Step 6.3; represents the relative membership degree of the k-th scheduling objective of the j-th feasible solution calculated in Step 6.2; and represent the optimal relative membership degree and the worst relative membership degree of the k-th scheduling objective among all feasible solutions respectively.

[0079] Calculate the relative membership degrees of all feasible solutions, and sort the solution set based on this. Select the feasible solution with the largest relative membership degree as the optimal solution under this decision-making preference, and recommend the corresponding reservoir optimal scheduling diagram and scheduling process to the reservoir.

[0080] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0081] Based on the traditional multi-objective optimal scheduling of reservoirs, the present invention adds the consideration of the ecological objectives of downstream lakes. By constructing a basin hydrological response model, it simulates the complex hydrological and ecological responses of downstream lakes to the upstream reservoir scheduling process, and couples it to the multi-objective optimal scheduling model of the river-lake system to achieve the common optimization of lake ecological objectives and reservoir beneficial objectives. It can, while ensuring the beneficial effects of upstream reservoirs, alleviate the adverse impacts of reservoir impoundment on the downstream lake ecology as much as possible; adopt the fuzzy optimization method to recommend the optimal scheduling rules under different decision-making preferences, providing a model basis for analyzing the benefit conversion mechanism between lake ecology and reservoir benefits, and providing scientific technical support and decision-making basis for the green development of the river-lake system. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 is a flow chart of an ecological scheduling method for a river-lake system coupling basin hydrological response provided by the present invention;

[0083] Figure 2 is a schematic diagram of the reservoir optimal scheduling rule;

[0084] Figure 3 is a topological diagram of the hydrological evolution from the Three Gorges Reservoir to the Dongting Lake Basin in the embodiment;

[0085] Figure 4 The scatter plot of the approximate Pareto solution set of the multi-objective optimal scheduling model in the embodiment; Figure 4 In (a), it is the target of the lake ecological water shortage degree and the target of ecological satisfaction degree of the approximate Pareto solution set; Figure 4 In (b), it is the target of the reservoir power generation and the target of the end-of-flood storage volume of the approximate Pareto solution set;

[0086] Figure 5 The optimal operation chart of the Three Gorges Reservoir recommended according to the decision-making preference in the embodiment. Specific implementation manners

[0087] The present invention will be further described below in conjunction with specific embodiments.

[0088] The present invention selects the area from the Three Gorges Reservoir in the Yangtze River Basin to the Dongting Lake Basin, and develops an ecological regulation method for the river-lake system that couples the basin hydrological response. The embodiment pays attention to the adverse ecological impacts caused by the concentrated water storage of the upstream reservoir group in the Yangtze River during the water storage period on the downstream Dongting Lake. Taking the improvement of the hydrological ecology of the Dongting Lake during the water storage period as the goal, taking into account the beneficial effects of the Three Gorges Reservoir, the power generation operation chart of the Three Gorges Reservoir is optimized, providing a scientific basis for the sustainable development of the river-lake system. The specific implementation manners will be described in detail in combination with the technical solution and the drawings, which specifically include the following steps:

[0089] Step 1: Select representative hydrological stations in the study area and draw the topological map of the basin hydrological evolution.

[0090] Step 1.1: Collect basin information and analyze characteristics. Based on the topographic map and water system map of the area between the Three Gorges Reservoir in the Yangtze River Basin and the Dongting Lake, clarify the distribution of the basin water system, the connection relationship of the river channels, and the water flow direction and path. Specifically:

[0091] The Three Gorges Reservoir is located in the upper reaches of the Yangtze River. Its discharged water evolves from west to east along the main channel of the Yangtze River, and is diverted in the Jingjiang River Basin. Part of the water volume flows into the Dongting Lake through the three outlets area, and the rest of the water volume continues to evolve downstream along the main channel of the Yangtze River.

[0092] The Dongting Lake is located in the middle reaches of the Yangtze River. Its inflowing runoff includes not only the inflow of the main channel of the Yangtze River in the three outlets area, but also the inflow of 4 tributaries in the four rivers area (Xiangjiang River, Zijiang River, Yuanjiang River, and Lishui River), and then flows out near the Chenglingji Hydrological Station and into the main channel of the Yangtze River. Based on the geographical characteristics, the lake area can be divided into three lake areas: East Dongting Lake, West Dongting Lake, and South Dongting Lake, and there are obvious differences in the hydrological response characteristics of different lake areas.

[0093] Step 1.2: Select representative hydrological stations in the basin. Based on the existing hydrological station network in the basin, select the hydrological stations with long-term, continuous, and consistent historical measured data as the representative stations for subsequent construction of the basin hydrological response model. Specifically include:

[0094] Step 1.2.1, reservoir representative stations: Set a reservoir outflow station at the dam site of the Three Gorges Reservoir.

[0095] Step 1.2.2, river channel representative stations: Set Zhicheng Station and Shashi Station in the Jingjiang area of the main stream of the Yangtze River, set Xinjiangkou Station, Shadaoguan Station, Mitosi Station, Kangjiagang Station and Guanjiapu Station on the five tributaries in the three-outlet area respectively, and set Shimen Station, Taoyuan Station, Taojiang Station and Xiangtan Station at the confluences of the four tributaries in the four-river area respectively.

[0096] Step 1.2.3, lake area representative stations: According to the spatial hydrological response characteristics within the lake, set representative stations in the three lake areas of Dongting Lake respectively, namely Chenglingji Station in the East Dongting Lake, Yangliutan Station in the South Dongting Lake, and Nanjui Station in the West Dongting Lake.

[0097] Step 1.3, draw the topological map of basin hydrological evolution. Generalize the basin water system structure, use the representative stations selected in Step 1.2 as nodes, use straight lines to represent the hydraulic connections between nodes, and use arrows to indicate the flow direction; use rectangles to mark the lake boundaries, and indicate the names of the reservoirs and each representative station within the study area, so as to form a topological map of basin hydrological evolution that can accurately reflect the spatio-temporal correlation between the reservoir, river and lake as Figure 3 shown.

[0098] Step 2, gradually construct a basin hydrological response model to simulate the response process of lake hydrological situation to the upstream reservoir discharge.

[0099] Step 2.1, decompose the study area and set hydrological response sub-models step by step. Based on the topological map of basin hydrological evolution drawn in Step 1.3, decompose the study area, set hydrological response sub-models for each group of adjacent representative stations respectively, and gradually quantify the basin hydrological response.

[0100] Step 2.1.1, for the reservoir representative stations and river channel representative stations, the hydrological response sub-model takes the water volume characteristics of the upstream representative station among the adjacent representative stations as the input variable and the water volume characteristics of the downstream representative station as the output variable, and constructs the hydrological response sub-model in turn through two steps of determining the function type and optimizing the parameter values. The hydrological characteristics of each representative station can select the water level, water depth, flow rate or water volume characteristics of the representative station.

[0101] Step 2.1.2, for the lake representative stations, since there are 3 lake representative stations in the study area of the embodiment, first select the Chenglingji representative station in the East Dongting Lake, and construct a hydrological response sub-model with the water volume characteristics of the river channel representative stations where the lake inflow is located (that is, the 5 representative stations in the three-outlet area and the 4 representative stations in the four-river area in Step 1.2.2); then take the water level of Chenglingji Station as the input and the water levels of other lake representative stations (that is, Yangliutan Station in the South Dongting Lake and Nanjui Station in the West Dongting Lake) as the output, and construct a hydrological response sub-model to quantify the hydrological correlation between multiple lake areas within the lake.

[0102] All 10 decomposed sub-models of the basin hydrological response are shown in Table 1.

[0103] Step 2.2: Determine the function type of the hydrological response sub-model.

[0104] Step 2.2.1: The candidate function types include, but are not limited to, linear functions, polynomial functions, multivariate polynomial functions, or piecewise functions.

[0105] Step 2.2.2: Use the 5-fold cross-validation method to train and test all candidate function types. Using the historical measured data of the hydrological characteristics of the selected representative stations as the data set, evenly divide it into 5 subsets. Each time, select one subset as the test set, and the remaining 4 subsets as the training set. Train and test each candidate function type.

[0106] Step 2.2.3: Evaluate and optimize the accuracy of the candidate function types. Select the Pearson correlation coefficient CC, root mean square error RMSE, and relative bias RB as statistical indicators, and quantify the simulation accuracy of each candidate function type with the average value of its 5 training and tests. Select the function type with the best simulation accuracy for the hydrological response sub-model.

[0107] Step 2.3: Optimize the parameter values of the hydrological response sub-model. Use all the data sets in Step 2.2.2 to train the function type selected in Step 2.2.3, and optimize and determine the parameter values of the hydrological response sub-model.

[0108] Step 2.4: Integrate the hydrological response sub-models to form a complete basin hydrological response model.

[0109] Based on the function type determined in Step 2.2 and the parameter values optimized in Step 2.3, determine the structure of the hydrological response sub-model, and integrate the hierarchically constructed hydrological response sub-models into a complete basin hydrological response model, as shown in Table 1.

[0110] When applying the basin hydrological response model, use the hydrological response sub-models step by step in the order from upstream to downstream. Use the output of the upstream sub-model as the input data for the downstream sub-model to simulate the lake water level at each time period for the upstream reservoir discharge and the step-by-step dynamic response of the inflow from the intermediate area This process can be characterized as:

[0111] (1)

[0112] Table 1 Hydrological Response Model of the River-Lake System in the Yangtze River Three Gorges - Dongting Lake Basin

[0113]

[0114] Step 3: Propose ecological evaluation indicators and quantify the lake's hydrological-ecological response.

[0115] In this embodiment, taking the ecological water demand process of Dongting Lake as a reference, the ecological benefits of Dongting Lake at different water levels are quantified from two aspects: ecological satisfaction degree and ecological water shortage degree, that is, the hydrological-ecological response of the lake.

[0116] Step 3.1: Quantify the ecological satisfaction degree of the lake.

[0117] Set ecological satisfaction indicators for the lake to measure whether the water levels in each lake area of the lake can meet the ecological water level requirements during each period; then, according to the importance weights of the lake areas, the ecological satisfaction of each lake area is weighted to obtain the ecological guarantee degree of the lake during the period. The formula is as follows:

[0118] (2)

[0119] (3)

[0120] Among them, represents the ecological satisfaction degree of the lake at time t, which is the weighted average of the ecological satisfaction of each lake area. represents the ecological satisfaction of the i-th lake area at time t, and the value is 0 or 1. When the simulated water level LV i,t of the representative station in the i-th lake area during the period is not lower than the ecological demand D i,t , it is considered that the lake area meets the ecological demand, and W i,t is taken as 1, otherwise 0. represents the importance weight of the lake area. The weights of the East Dongting, West Dongting, and South Dongting are taken as 0.5, 0.2, and 0.3 respectively, and N zone represents the three lake areas of Dongting Lake.

[0121] Step 3.2: Quantify the ecological water shortage degree of the lake.

[0122] Taking the square of the ratio of the ecological water level to the simulated water level of the lake area as the ecological water shortage degree indicator to measure the ecological water shortage degree of each lake area during the period. If the lake area is not ecologically water short, the indicator value is in the interval (0,1]; otherwise, if the lake area is ecologically water short, the indicator is greater than 1, and the closer the value is to positive infinity, the lower the simulated water level during the period and the more serious the water shortage degree. The formula is as follows:

[0123] (4)

[0124] (5)

[0125] Among them, represents the ecological water shortage degree of the lake at time t, which is the weighted average of the ecological water shortage degree of each lake area. represents the ecological water shortage degree of the \(i\)-th lake area in the \(t\) period, and the value range is \((0, +\infty)\).

[0126] Step 4: Construct a multi-objective optimal operation model for the river-lake system.

[0127] Step 4.1: Reservoir optimal operation chart and decision variable setting.

[0128] Based on the conventional reservoir operation rules, set the reservoir optimal operation chart in the multi-objective optimal operation. The form is as Figure 2 shown. The reservoir optimal operation chart includes three operation lines, namely the reduced output line, the guaranteed output line, and the increased output line. The three output lines divide the operation interval into four output areas, namely the reduced output area, the guaranteed output area, the increased output area 1, and the increased output area 2. The output coefficients of each output area are , , and , satisfying the constraint . When conducting reservoir power generation operation based on the operation chart, the output in each period is determined by the output area where the water level at the beginning of the period is located, and the output size is equal to the product of the output coefficient and the guaranteed output. Referring to the conventional operation rules of the Three Gorges Reservoir, the optimal operation rules in this embodiment adopt a ten-day scale, that is, there are 36 periods in a year.

[0129] Therefore, the decision variables of the optimal operation chart in this embodiment include: 108 water level variables of 3 operation lines * 36 operation periods, and 3 output coefficient variables other than the guaranteed output area , and .

[0130] Step 4.2: Set the objective function. The operation objectives include the lake ecological objective and the reservoir beneficial operation objective. Among them, the lake ecological objective is the maximization objective of lake ecological satisfaction and the minimization objective of lake ecological water shortage degree, and the reservoir beneficial operation objective is the maximization objective of the dimensionless reservoir power generation and the maximization objective of the end-of-flood season water storage. The definitions and formulas of each objective are as follows:

[0131] Step 4.2.1: Maximization objective of lake ecological satisfaction. To achieve the maximization of lake ecological satisfaction, the lake ecological satisfaction in each period in Step 3.1 is weighted by period. The period importance weights are allocated according to the current ecological situation of the basin and the basin operation requirements. Specifically, for the critical periods with more frequent ecological water shortage, more severe degree, and more concerned operation requirements, higher weights should be assigned; while for the periods that can better meet the ecological needs in the current situation, the importance weights should be appropriately reduced, or even can be set to 0. The formula is as follows:

[0132] (6)

[0133] Among them, is the weighted lake ecological satisfaction degree for each scheduling period, and are the overall lake ecological satisfaction degree and the ecological satisfaction of the i-th lake area at time t, respectively. is the importance weight of the period. Since this embodiment focuses on the ecological response of Dongting Lake during the reservoir impoundment period (September and October), the weights for the first, middle, and last ten-day periods in September and October are taken as 1 / (6 ten-day periods * 60 years) = 1 / 360, and the weights for the scheduling periods in other months are taken as 0; N period represents the number of optimized scheduling periods, which is 36 ten-day periods * 60 years = 2160.

[0134] Step 4.2.2, the goal of minimizing the lake ecological water shortage degree: Weight the lake ecological water shortage degree indicators for each period in Step 3.2 by period.

[0135] (7)

[0136] Among them, is the weighted lake ecological water shortage degree for each scheduling period, and are the overall ecological water shortage degree of the lake and the ecological water shortage degree of the i-th lake area at time t, respectively.

[0137] Step 4.2.3, the goal of maximizing the reservoir power generation: The average annual power generation over the years is maximized, and the goal is made dimensionless based on the initial design or the average annual power generation statistically obtained from historical data.

[0138] (8)

[0139] Among them, is the dimensionless average annual power generation of the reservoir over the years; N output is the guaranteed output of the Three Gorges Reservoir, taken as 499 kW, is the duration of the scheduling period, that is, one ten-day period, is the output coefficient of the period; is the average annual power generation, taking the designed annual power generation of the Three Gorges Reservoir as 84.7 billion kWh, N year is the total number of years participating in the optimized scheduling, which is 60 years.

[0140] Step 4.2.4, maximizing the reservoir's end-of-flood-season water storage: The average end-of-flood-season water storage of the reservoir over the years is maximized, and it is characterized by the average annual storage capacity at the end of the reservoir impoundment period.

[0141] (9)

[0142] Among them, is the dimensionless average annual end-of-flood-season water storage; Vn is the reservoir capacity at the end of the water storage period in the nth year. In this embodiment, the reservoir capacity on November 1 of the Three Gorges Reservoir is taken as an example, V max is the total reservoir capacity of the Three Gorges Reservoir, taking 39.3 billion m 3 .

[0143] Step 4.3, set the constraint conditions. The constraints of the multi-objective optimal operation model include: water balance constraint, reservoir water level constraint, reservoir discharge flow constraint, hydropower station output constraint, and non-negativity constraint of variables.

[0144] Step 5, optimize the multi-objective optimal operation model of the river-lake system.

[0145] The optimization and solution of the multi-objective optimal operation model follow the general idea of "simulating the operation process - optimizing the operation rules". This embodiment is carried out based on the epsilon-NSGA-II algorithm.

[0146] Step 5.1, simulate the operation process. Based on the values of the decision variables, determine the current reservoir optimal operation chart. Taking the inflow runoff process of the Three Gorges Reservoir as the input, conduct water balance calculations to derive the reservoir water level, discharge flow, and output process. Input the reservoir discharge process into the basin hydrological response model constructed in Step 2 to derive the spatial water level response process of Dongting Lake, and then evaluate the lake ecological level based on the indicators proposed in Step 3, thereby deriving the values of the lake ecological target and the reservoir beneficial operation target under the current operation chart, and returning them to the optimization stage.

[0147] Step 5.2, optimize the operation rules. Based on the selected optimization algorithm, generate an initial solution set; evaluate and screen the solution set based on the objective values calculated in the operation process simulation stage of Step 5.1; through iterative optimization, gradually improve the solution set until the number of iterations is equal to the preset threshold; finally, output the preferred approximate Pareto solution set, as well as the corresponding reservoir operation process and lake response process. Figure 4 In (a) shows the lake ecological water shortage degree target and ecological satisfaction degree target of the preferred approximate Pareto solution set of the embodiment, Figure 4 In (b) shows the reservoir power generation target and the end-of-flood-season water storage target of the approximate Pareto solution set.

[0148] Step 6, based on the fuzzy optimization method, recommend the optimal reservoir operation rules that meet the decision-making preferences.

[0149] Step 6.1, calculate the improvement ability of the optimal operation chart compared with the conventional operation. Statistically calculate the optimal values and average values of the operation targets and important operation characteristics of all feasible solutions in the approximate Pareto solution set. Compared with the conventional reservoir operation, calculate the relative percentage increase in each operation target as shown in Table 2.

[0150] Table 2 Comparison of operation targets and characteristics between the optimal operation rules and the conventional operation rules

[0151]

[0152] Step 6.2, calculate the relative membership degree of the scheduling objectives of the Pareto solution set. Since the value ranges of various scheduling objectives are different, they are standardized separately, and the relative membership degree of the scheduling objectives of the Pareto solution set is obtained by using formula (10). In the embodiment, the greater the satisfaction degree of the lake ecosystem, the power generation of the reservoir, and the storage capacity of the reservoir at the end of the flood season, the better, and the smaller the water shortage degree of the lake ecosystem, the better.

[0153] (10)

[0154] Wherein, and are respectively the objective value and the relative membership degree of the i-th scheduling objective of the j-th feasible solution, and are respectively the minimum value and the maximum value of the i-th scheduling objective.

[0155] Step 6.3, calculate the weights of the scheduling objectives according to the decision-making preference. According to the decision-making preference, the importance of each pair of objectives is compared. The fuzzy tone operators such as "equally important, slightly important, moderately important, relatively important, significantly important, highly important, very important, extremely important, absolutely important, incomparable" are used to determine the fuzzy scale value and describe the relative importance degree between the objectives. Among them, the fuzzy scale values of equally important and incomparable are 0.5 and 1.0 respectively, and the fuzzy scale values of other fuzzy tones are calculated by linear interpolation method. In this embodiment, it is considered that the satisfaction degree of the lake ecosystem is the most important, equally important with the water shortage degree of the lake ecosystem, and significantly more important than the power generation of the reservoir and the storage capacity of the reservoir at the end of the flood season. Therefore, the fuzzy expected importance degrees of the four objectives are (0.5, 0.5, 0.75, 0.75).

[0156] Based on the given fuzzy tone operators, substituting them into formula (11) to obtain the importance degrees of different objectives, and after normalizing them, the weights of each scheduling objective are =(0.375, 0.375, 0.125, 0.125).

[0157] (11)

[0158] Wherein, represents the importance degree of all scheduling objectives; , , and respectively represent the importance degrees of objective 1 to objective 4, where objective 1 is the most important objective considered by the decision maker; , and They respectively represent the fuzzy scale values of the 2nd, 3rd, and 4th targets compared with the target 1, which are determined by the fuzzy tone operator determined by the decision maker.

[0159] Step 6.4 Calculate the relative membership degree and recommend the optimal scheduling diagram that meets the decision preference. Assume that the relative membership degrees of the optimal solution and the worst solution are and respectively. Then the relative membership degree of the jth feasible solution with respect to the optimal solution is:

[0160] (12)

[0161] where represents the relative membership degree of the jth feasible solution; represents the weight of the kth scheduling target calculated in step 6.3; represents the relative membership degree of the kth scheduling target of the jth feasible solution calculated in step 6.2; and respectively represent the optimal relative membership degree and the worst relative membership degree of the kth scheduling target among all feasible solutions.

[0162] In the embodiment, the maximum value of the relative membership degree is 0.830, and the target relative membership degrees of the corresponding feasible solution are (0.8571, 0.7825, 0.8884, 0.8333). Its scheduling diagram is as Figure 5 shown, which is the optimal reservoir scheduling diagram recommended under this decision preference.

[0163] The above embodiments only represent the implementation manners of the present invention, but should not be construed as limiting the scope of the patent of the present invention. It should be noted that for those skilled in the art, without departing from the concept of the present invention, several deformations and improvements can still be made, and these all belong to the protection scope of the present invention.

Claims

1. A method for ecological regulation of river-lake systems coupled with basin hydrological response, characterized in that: The river-lake system ecological dispatching method comprises the following steps: Step 1: Select representative hydrological stations in the study area and draw a topological map of the basin hydrological evolution; Step 2: construct a watershed hydrological response model step by step to simulate the response of the lake hydrological regime to the discharge of the upstream reservoir; Step 3, propose ecological assessment indicators to quantify the lake hydrological-ecological response; The dynamic response process of lake water level is simulated by basin hydrological response model. Based on the ecological water demand process of lake, the ecological benefits of lake under different water levels are quantified from two aspects: ecological satisfaction and ecological water shortage, and the hydrological-ecological response of lake is obtained. Step 4: Construct a multi-objective optimization scheduling model for river and lake systems; Step 5, optimizing the multi-objective optimization scheduling model of the river-lake system; the optimization and solution of the multi-objective optimization scheduling model follow the overall idea of ​​"scheduling process simulation-scheduling rule optimization"; Step 6: Based on the fuzzy optimization method, the optimal reservoir operation rules that meet the decision-making preferences are recommended; Calculate the improvement capability of the optimized scheduling diagram compared with the conventional scheduling, calculate the relative membership of the scheduling target of the Pareto solution set, calculate the relative superiority of the scheduling target weight according to the decision preference, and recommend the optimal scheduling diagram that meets the decision preference; The specific implementation of step 3 is as follows: Step 3.1, quantify the ecological satisfaction of the lake; The lake ecological satisfaction index is set to measure whether the water level of each lake area in the lake at each time period can meet the ecological water level demand; then the ecological satisfaction of each lake area is weighted according to the importance weight of the lake area to obtain the ecological guarantee degree of the lake at that time period; the formula is as follows: Among them, W t lake It represents the ecological satisfaction of the lake in period t, which is the weighted average of the ecological satisfaction of each lake area; Indicates the ecological satisfaction of the i-th lake area in period t, with a value of 0 or 1; the simulated water level LV of the representative station of the i-th lake area in that period i,t Not less than ecological demand D i,t When it is believed that the lake area meets the ecological needs, W i,t is 1, otherwise it is 0; represents the importance weight of lake area i, determined based on expert experience, N zone Indicates the number of partitions contained in the lake, satisfying and Constraints; Step 3.2, quantify the degree of lake ecological water shortage; The square of the ratio of the lake ecological water level to the simulated water level is used as the ecological water shortage index to measure the ecological water shortage degree of each lake area in the period; if the lake area is not short of water, the index value is between (0,1]; on the contrary, if the lake area is short of water, the index is greater than 1, and the closer the value is to positive infinity, the lower the simulated water level is during the period, and the more serious the water shortage is; the formula is as follows: in, It represents the ecological water shortage degree of the lake in period t, which is the weighted average of the ecological water shortage degree of each lake area; It represents the ecological water shortage degree of the i-th lake area in period t, and its value range is (0, +∞); The specific implementation of step 4 is as follows: Step 4.1, reservoir optimization operation diagram and decision variable setting; Based on the conventional reservoir dispatching rules, a reservoir optimization dispatching diagram in multi-objective optimization dispatching is set; the reservoir optimization dispatching diagram includes three dispatching lines, namely, the output reduction line, the output guarantee line and the output increase line; the three output lines divide the dispatching interval into four output areas, namely, the output reduction area, the output guarantee area, the output increase area 1 and the output increase area 2; the output coefficients of each output area are respectively and Satisfy constraints When the reservoir power generation is dispatched based on the dispatch diagram, the output of each period is determined by the output area where the initial water level of the period is located, and the output is equal to the product of the output coefficient and the guaranteed output; The reservoir optimization dispatching diagram includes two types of decision variables, namely, the water level variable in each dispatching period that determines the output line position, and the output coefficient variable that determines the output size in each period; Step 4.2, set the objective function; the dispatching objectives include lake ecological objectives and reservoir benefit objectives; among them, the lake ecological objectives are the objectives of maximizing the lake ecological satisfaction and minimizing the lake ecological water shortage, and the reservoir benefit objectives are the dimensionless objectives of maximizing the reservoir power generation and maximizing the water storage at the end of the flood season; Step 4.3, set constraints; the constraints of the multi-objective optimization scheduling model include: water balance constraints, reservoir water level constraints, reservoir discharge flow constraints, hydropower station output constraints, and variable non-negative constraints; In step 4.2, the definitions and formulas of each target are as follows: Step 4.2.1, the goal of maximizing the degree of lake ecological satisfaction; in order to maximize the degree of lake ecological satisfaction, the degree of lake ecological satisfaction in each period in step 3.1 is weighted; the importance weight of the period is allocated according to the ecological status of the basin and the basin scheduling needs; specifically, for key periods with more frequent ecological water shortages, more severe levels, and more attention to scheduling needs, higher weights should be given; and for periods that can better meet ecological needs in the current situation, the importance weight is appropriately reduced, and can even be set to 0; the formula is as follows: Among them, F eco-reliability is the weighted lake ecological satisfaction degree in each dispatching period, W t lake and They are the overall ecological satisfaction of the lake and the ecological satisfaction of the i-th lake area in period t respectively; is the importance weight of the time period; N period is the number of scheduling periods, satisfying and is the importance weight of lake region i, N zone is the number of lake partitions that satisfy and Step 4.2.2, minimizing the ecological water shortage degree of the lake: weighting the ecological water shortage degree index of each period in step 3.2; Among them, F eco-shortage is the weighted ecological water shortage degree of the lake in each dispatching period, and are the overall ecological water shortage degree of the lake and the ecological water shortage degree of the i-th lake area in period t respectively; Step 4.2.3, the goal of maximizing reservoir power generation: the average power generation over many years is maximized, and the goal is dimensionless based on the annual average power generation of the initial design or historical data statistics; Among them, F power is the dimensionless average power generation of the reservoir over many years; N output is the guaranteed output of the reservoir, △t is the duration of the dispatching period, is the output coefficient of the time period, which is determined by the output area where the initial water level of the time period is located, and its value is or is the average annual power generation, N year is the scheduling year number; Step 4.2.4, maximizing the reservoir storage at the end of the flood season: the multi-year average end-of-flood storage of the reservoir is the largest, represented by the multi-year average storage capacity at the end of the reservoir storage period; Among them, F storage V is the dimensionless multi-year average flood season water storage capacity; n is the reservoir capacity at the end of the storage period in the nth year, V max is the total storage capacity of the reservoir.

2. The ecological dispatching method of river-lake system coupled with basin hydrological response according to claim 1 is characterized in that: The specific implementation of step 1 is as follows: Step 1.1, collect watershed information and analyze characteristics; Step 1.2, select representative hydrological stations in the basin; based on the existing hydrological station network in the basin, select hydrological stations with long-term, continuous and consistent historical measured data as representative stations for the subsequent construction of the basin hydrological response model; Step 1.3, draw a topological map of the hydrological evolution of the basin; generalize the water system structure of the basin, use the representative stations selected in step 1.2 as nodes, use straight lines to represent the hydraulic connection between nodes, and use arrows to indicate the flow direction; use rectangles to mark the boundaries of lakes, and indicate the names of reservoirs and representative stations in the study area to form a topological map of the hydrological evolution of the basin.

3. The ecological dispatching method of river-lake system coupled with basin hydrological response according to claim 2 is characterized in that: In step 1.2, the selection of representative stations should follow the following principles: Step 1.2.1, Reservoir representative station: A reservoir outflow station should be set up at the reservoir dam site; Step 1.2.2, River representative station: At least one representative station should be selected for each main and tributary river; specifically, in the river section where tributaries flow into or diverge, representative stations should be set up both upstream and downstream of the intersection; In river sections where there are no tributaries or diversions, a representative station is set up, and additional representative stations are added in cases where the river terrain is complex; Step 1.2.3, lake representative station: at least one representative station should be selected inside the lake area, and the specific number is determined according to the spatial hydrological response characteristics inside the lake; if there is no obvious difference in the spatial hydrological response inside the lake, then one lake representative station can be selected; if there is a significant difference in the spatial hydrological response inside the lake, then it will be divided into multiple lake areas according to the lake's topographic characteristics and hydrological response characteristics, and a representative station will be set up in each lake area.

4. The ecological dispatching method of river-lake system coupled with basin hydrological response according to claim 2 is characterized in that: The specific implementation of step 2 is as follows: Step 2.1, decompose the study area and set up the hydrological response sub-model step by step; based on the basin hydrological evolution topology map obtained in step 1.3, decompose the study area, set up the hydrological response sub-model for each group of adjacent representative stations, and quantify the basin hydrological response step by step; Step 2.2, determine the function type of the hydrological response sub-model; Step 2.2.1, the candidate function types include linear function, polynomial function, multivariate polynomial function or piecewise function; Step 2.2.2, use the k-fold cross-validation method to train and test all candidate function types; use the historical measured data of the hydrological characteristics of the selected representative station as the data set, divide it evenly into k subsets, select one subset as the test set each time, and the remaining k-1 subsets as the training set, and train and test each candidate function type; Step 2.2.3, evaluate the accuracy and optimize the candidate function types; select the Pearson correlation coefficient CC, root mean square error RMSE and relative deviation RB as statistical indicators, quantify the simulation accuracy of each candidate function type by the average value of k training tests, and select the function type with the best simulation accuracy for the hydrological response submodel; Step 2.3, optimizing the parameter values ​​of the hydrological response sub-model; using all the data sets in step 2.2.2 to train the function type selected in step 2.2.3, and optimizing and determining the parameter values ​​of the hydrological response sub-model; Step 2.4, integrating the hydrological response sub-models to form a complete watershed hydrological response model; Based on the function type determined in step 2.2 and the parameter values ​​optimized in step 2.3, determine the structure of the hydrological response sub-model, and integrate the hierarchically constructed hydrological response sub-models into a complete watershed hydrological response model; When applying the basin hydrological response model, the hydrological response sub-models are applied step by step from upstream to downstream, and the output of the upstream sub-model is used as the input data of the downstream sub-model to simulate the lake water level LV at each time period. t Discharge to upstream reservoirs and interval inflow The step-by-step dynamic response of the process can be characterized as follows:

5. The method for ecological dispatching of river-lake system coupled with basin hydrological response according to claim 4, characterized in that: The step 2.1 is specifically as follows: Step 2.1.1, for the reservoir representative station and the river representative station, the hydrological response sub-model takes the hydrological characteristics of the upstream representative station in the neighboring representative station as the input variable and the hydrological characteristics of the downstream representative station as the output variable, and constructs the hydrological response sub-model in sequence through the two steps of determining the function type and optimizing the parameter value; Hydrological characteristics of each representative station Select the water level, water depth, flow or water volume characteristics of the representative station; Step 2.1.2: For lake representative stations, when there is only one lake representative station in the study area, the hydrological characteristics of the river representative station flowing into the lake are used as input variables and the hydrological characteristics of the lake representative station are used as output variables to construct a lake hydrological response sub-model to quantify the response of the lake hydrological regime to the runoff into the lake; when there are multiple lake representative stations in the study area, one of the lake representative stations is first selected to construct a hydrological response sub-model with the river representative station flowing into the lake; Then, taking the hydrological characteristics of the selected lake representative station as input and the hydrological characteristics of other lake representative stations as output, a hydrological response sub-model is constructed to quantify the hydrological correlation between multiple lake areas within the lake.

6. The ecological dispatching method of river-lake system coupled with basin hydrological response according to claim 1 is characterized in that: The specific implementation of step 5 is as follows: Step 5.1, dispatching process simulation; based on the decision variable values, determine the current reservoir optimization dispatching diagram, take the reservoir runoff process as input, perform water balance calculation, and deduce the reservoir water level, discharge flow and output process; input the reservoir discharge process into the basin hydrological response model constructed in step 2, deduce the spatial water level response process of the downstream lake, and then evaluate the lake ecological level based on the indicators proposed in step 3, thereby deduce the lake ecological target and reservoir benefit target values ​​under the current dispatching diagram, and return them to the optimization stage; Step 5.2, scheduling rule optimization: generating an initial solution set based on the selected optimization algorithm; Evaluate and screen the solution set based on the target value calculated in step 5.1 during the simulation phase of the scheduling process; Through iterative optimization, the solution set is gradually improved until the number of iterations is equal to the preset threshold or the target value variation is less than the preset threshold; finally, the preferred approximate Pareto solution set and the corresponding reservoir scheduling process and lake response process are output; the approximate Pareto solution set includes one or more feasible solutions, each feasible solution corresponds to a set of values ​​of the decision variables described in step 4.1, based on which the reservoir optimization scheduling diagram corresponding to the current feasible solution can be drawn.

7. The ecological dispatching method of river-lake system coupled with basin hydrological response according to claim 6 is characterized in that: The specific implementation of step 6 is as follows: Step 6.1, calculate the improvement capacity of the optimized dispatching diagram compared with the conventional dispatching; statistically calculate the optimal values ​​and average values ​​of the dispatching objectives and important dispatching characteristics of all feasible solutions in the approximate Pareto solution set; and calculate the relative percentage of improvement of each dispatching objective compared with the conventional dispatching of the reservoir; Step 6.2, calculate the relative membership of the scheduling target of the Pareto solution set; the value ranges of various scheduling targets are different, so they are standardized respectively, and the relative membership of the scheduling target of the Pareto solution set is obtained using formula (10); Among them, f i,j and r i,j are the target value and relative membership of the ith scheduling objective of the jth feasible solution, respectively, and f i,min and f i,max are the minimum and maximum values ​​of the i-th scheduling objective respectively; Step 6.3, calculate the dispatch target weight according to the decision preference; according to the decision preference, compare the importance of each pair of targets, use "equally important, slightly important, slightly important, relatively important, obviously important, significantly important, very important, very important, extremely important, extremely important, incomparable" as the fuzzy mood operator, determine the fuzzy scale value, and describe the relative importance of the targets; among them, the fuzzy scale values ​​of equally important and incomparable are 0.5 and 1.0 respectively, and the fuzzy scale values ​​of other fuzzy tones are calculated using linear interpolation; based on the assigned fuzzy mood operator, substitute it into formula (11) to obtain the importance of different targets, and after normalization, obtain the weight of each dispatch target ω=(ω1,ω2,ω3,ω4); Among them, ω′ represents the importance of all scheduling objectives; ω1′, ω2′, ω3′ and ω4′ represent the importance of objectives 1 to 4 respectively, where objective 1 is the most important objective considered by the decision maker; u 12 、u 13 and u 14 They represent the fuzzy scale values ​​of the second, third, and fourth objectives, respectively, compared with objective 1, which are determined by the fuzzy mood operator determined by the decision maker; Step 6.4, calculate the relative superiority and recommend the optimal scheduling diagram that meets the decision preference; suppose the relative membership of the optimal solution and the worst solution are g = (g1, g2, g3, g4) and l = (l1, l2, l3, l4), respectively, then the relative superiority of the jth feasible solution relative to the optimal solution is: Among them, u g,j represents the relative superiority of the jth feasible solution; ω k represents the weight of the kth scheduling target calculated in step 6.3; r k,j represents the relative membership of the kth scheduling objective of the jth feasible solution calculated in step 6.2; g k and l k They represent the optimal relative membership and the worst relative membership of the kth scheduling objective in all feasible solutions respectively; The relative superiority of all feasible solutions is calculated, and the solution set is sorted based on this. The feasible solution with the largest relative superiority is taken as the optimal solution under this decision preference, and the corresponding reservoir optimization scheduling diagram and scheduling process are recommended to the reservoir.

Images