A water quantity scheduling risk decision method for inter-basin water diversion projects

Abstract

The application discloses a kind of interbasin water transfer engineering water quantity scheduling risk decision-making methods, comprising: building interbasin water transfer system decennium optimization scheduling model, to decennium generation scheduling scheme;According to decennium optimization scheduling model, the threshold value of tolerable unit water transfer energy consumption is calculated, and when unit water transfer energy consumption exceeds the threshold value, then unit water transfer energy consumption risk occurs;While considering water transfer task risk, establish the evaluation criteria of water transfer risk occurrence;Based on the evaluation criteria, improve decennium optimization scheduling model, consider inflow uncertainty, to water source inflow sequence and decennium water level stochastic simulation model generated water level sequence simulation value along water transfer line each lake water level sequence along water transfer line provide water level calculation boundary, build interbasin scheduling risk quantification and optimization decision-making model;Interbasin optimization scheduling scheme is obtained, and scheduling scheme risk and key risk factor sensitivity are evaluated.The method of the application clearly benefits and risk changes in interbasin water transfer process, provides technical support for safe and efficient operation of interbasin water transfer.

Description

Technical Field

[0001] This invention belongs to the field of inter-basin water transfer research, and more specifically, relates to a risk decision-making method for water allocation in inter-basin water transfer projects. Background Technology

[0002] The current research methods for cross-basin scheduling mainly involve generalizing and reducing the complexity of cross-basin scheduling systems, with the goal of optimizing water resource allocation. The objectives include minimizing water transfer costs, maximizing water storage satisfaction, and minimizing system pumping volume. Dynamic programming algorithms, genetic algorithms, and improved particle swarm optimization algorithms are used to solve the optimal scheme for cross-basin water volume scheduling. At the same time, the research uses risk assessment indicators such as water transfer guarantee rate, water supply reliability, and comprehensive risk level to conduct risk assessment on the obtained cross-basin water volume optimization scheduling scheme. However, these research methods still have the following two problems: (1) These studies are mostly based on historical deterministic water inflow processes, considering the adverse effects caused by water shortage in the water-receiving area or the impact of changes in the hydrological situation of the water-receiving area on the scheduling process, but neglecting the water supply capacity under the influence of uncertain water inflow from the water source area. The water inflow situation in the water source area is an important determining factor in cross-basin water transfer schemes. Uncertainty in water inflow will affect the stability of the cross-basin scheduling process and the feasibility of the scheduling scheme. (2) The optimization objectives of cross-basin water allocation do not consider risk values. That is, the current research methods independently consider cross-basin water allocation optimization and risk assessment. Due to the uncertainty and complexity of hydrological processes, there may be a situation where the optimization scheme and the allocation risk are not in harmony. Therefore, the current research methods for cross-basin allocation still have significant limitations in practical operation guidance. Summary of the Invention

[0003] Purpose of the invention: The purpose of this invention is to overcome the shortcomings of the prior art and provide a risk decision-making method for water allocation in inter-basin water transfer projects. Based on this method, the changes in benefits and risks in the process of inter-basin water transfer are derived, and a low-risk, optimized inter-basin water allocation scheme is provided.

[0004] Technical solution: The present invention provides a risk decision-making method for water allocation in inter-basin water transfer projects, comprising the following steps:

[0005] Spatially generalize the relationships between various engineering units within the inter-basin water transfer system, dividing the system into sections with lakes as nodes; and establish scheduling simulation models for each engineering unit. Based on these simulation models, establish hydraulic connections between the engineering units of the water transfer system. With the goal of minimizing the total operating cost of the pumping stations in the water transfer system, construct a ten-day optimal scheduling model for the inter-basin water transfer system. Use the actual water level of the lakes as the deterministic water level calculation boundary for each section, and make decisions on the ten-day scheduling scheme through the selection and switching of water sources and routes.

[0006] The tolerable unit water transfer energy consumption threshold is calculated based on the ten-day optimization scheduling model. When the actual unit water transfer energy consumption exceeds this threshold, it is considered that a unit water transfer energy consumption risk has occurred. At the same time, the risk of water transfer tasks is considered, and a judgment standard for the occurrence of water transfer risk is established.

[0007] Based on the criteria for assessing the occurrence of water transfer risks, the ten-day optimization scheduling model is improved. Taking into account the uncertainty of water inflow, the simulated values ​​of water level sequences of each lake along the water transfer route generated by the ten-day water level stochastic simulation model are used to provide the water level calculation boundary, and a cross-basin scheduling risk quantification and optimization decision-making model is constructed.

[0008] Based on the risk quantification and optimization decision-making model of cross-basin scheduling, a risk decision-making scheduling scheme for cross-basin water transfer system is obtained, and the risk of the scheduling scheme and the sensitivity of key risk factors are evaluated.

[0009] Furthermore, the objective function of the ten-day optimal scheduling model for the inter-basin water transfer system is:

[0010]

[0011] Among them, E year t represents the total operating cost of the pumping station; t represents the time period number, and T represents the total number of time periods; i n N0 represents the number of pump stations; For number i n The power consumption of the pump station during time period t;

[0012] The constraints of the model include: total water diversion volume constraint, lake water balance constraint, pumping station working capacity constraint, lake storage capacity constraint, pumping control water level constraint, river water conveyance capacity constraint, river water level constraint, and river water balance constraint.

[0013] Furthermore, the criteria for assessing the occurrence of water diversion risks are as follows:

[0014] A stochastic simulation model of the water level of lakes along the water transfer route is established based on the Copula function. The simulated values ​​of the lake water level sequence for each ten-day period are simulated. The simulated values ​​are then input into the ten-day optimal scheduling model of the inter-basin water transfer system along with different total water transfer task amounts. Multiple scheduling processes are simulated, and multiple ten-day scheduling schemes and their total unit water transfer energy consumption simulation values ​​are generated.

[0015] Determine the maximum water conveyance capacity of the river channel and pumping station, and determine the ten-day operation time of each section of the water transfer system according to the ten-day scheduling plan; combine the operation time and the maximum water conveyance capacity to determine the actual water transfer volume of each section in different ten-day scheduling plans;

[0016] Fisher's optimal segmentation method was used to divide the simulated values ​​of different total unit water transfer energy consumption into acceptable, tolerable and unacceptable zones. The upper and lower limits of unit water transfer energy consumption in each zone were calculated, and the upper limit of the tolerable zone was the tolerable unit water transfer energy consumption threshold.

[0017] When the actual water transfer volume during a given period is less than the water transfer task volume during that period, it is considered that a water transfer task risk has occurred; when the actual unit water transfer energy consumption is greater than the tolerable unit water transfer energy consumption threshold, it is considered that a unit water transfer energy consumption risk has occurred; when either the water transfer task risk or the unit water transfer energy consumption risk occurs, it is considered that a water transfer risk has occurred.

[0018] Furthermore, the method for constructing the risk quantification and optimization decision-making model for inter-basin water transfer is as follows: Based on the objectives and influencing conditions of water transfer, the inter-basin water transfer system is spatially generalized and divided into sections; for the scheduling and operation rules and patterns of lakes, rivers, pumping stations, and sluice gates, scheduling simulation models for each engineering unit are established; considering the uncertainty of inflow, the scheduling period is divided into K0 stages, and n is generated by the stochastic simulation model of the ten-day water level. k The group of water source water levels provides water level calculation boundaries for each section. With the objective function of minimizing the stage water transfer risk rate of the water transfer system, a cross-basin scheduling risk quantification and optimization decision-making model is constructed. The constraints of the model include: total water transfer volume constraint, lake water balance constraint, pump station working capacity constraint, lake storage capacity constraint, pumping control water level constraint, river water conveyance capacity constraint, river water level constraint, and river water balance constraint.

[0019] Furthermore, the solution method for the cross-basin scheduling risk quantification and optimization decision-making model is as follows: A reverse-order decision-making process is adopted, iteratively simulating the scheduling scenarios at each stage to solve the model. The last stage K0 within the scheduling period is taken as the first stage of the decision-making process, and n is generated using a stochastic water level simulation model. k The simulated water level series of lakes were used to calculate the water diversion risk rate corresponding to each set of simulated water level series. The water diversion task volume with the lowest water diversion risk rate was selected as the stage water diversion volume. The calculation was repeated for the previous stage to obtain the water diversion volume, corresponding water diversion risk rate, and benefits for each stage. The overall water diversion risk rate (Risk) of the risk decision-making and scheduling scheme was calculated. w The calculation formula is:

[0020]

[0021] in, For the kth water diversion system i Water transfer risk rate at each stage; P(n) k W num W task C num C threshold) represents the probability function of the occurrence of water diversion risk; num represents the group number of the simulated lake water level sequence used for the water level calculation boundary; W num W represents the actual water diversion volume when the simulated water level sequence of the num-th lake is used as the boundary for water level calculation; task C represents the total water transfer task. num C represents the actual unit energy consumption for water diversion when the simulated water level sequence of the num-th lake is used as the boundary for water level calculation; threshold The threshold for tolerable unit water transfer energy consumption.

[0022] Furthermore, the risk assessment method for the scheduling scheme is as follows: different total water transfer tasks, water transfer start times, and first-stage water transfer volumes are set as boundary conditions. The optimal solution is obtained based on the cross-basin scheduling risk quantification and optimization decision-making model, generating water transfer costs and water transfer risk rates for different risk decision scheduling schemes. Water transfer costs are considered as benefits, and the process of water transfer risk rate and benefit changes is obtained. From the perspectives of water transfer start time, first-stage water transfer volume, and main water source utilization of different risk decision scheduling schemes, the impact of these factors on the relationship between water transfer risk rate and benefits is analyzed, providing a basis for weighing the allocation of water transfer volume in different time periods during the operation of the cross-basin water transfer system.

[0023] Based on historical scheduling data, under a given total water transfer task, the scheduling risk rate and benefit changes of the risk decision scheduling scheme are obtained by comparing and analyzing the actual historical scheduling schemes with the cross-basin scheduling risk quantification and optimization decision-making model, and the impact of the previous decision on the subsequent process is analyzed.

[0024] Furthermore, the sensitivity assessment method for key risk factors is as follows: First, a total water transfer task and simulated values ​​of lake water level sequences along the route are selected as benchmark conditions, and the risk rate of the benchmark scheme is obtained by the cross-basin scheduling risk quantification and optimization decision-making model; then, the sensitivity coefficient of each key risk factor is obtained based on the change in water transfer risk rate caused by the change of each key risk factor.

[0025] Key risk factors include the total water diversion volume, the inflow and outflow of water to lakes, and the controlled water level of the regulating lakes; the formula for calculating the sensitivity coefficient is:

[0026]

[0027] Where S is the sensitivity coefficient, ΔRisk w ΔF represents the percentage change in the risk rate, where ΔF is the percentage change in a specific key risk factor.

[0028] The system corresponding to the method includes:

[0029] The ten-day-scale scheduling scheme generation unit is used to spatially generalize the relationships between various engineering units within the inter-basin water transfer system, dividing the system into sections with lakes as nodes; and establishing scheduling simulation models for each engineering unit. Based on the scheduling simulation models of each engineering unit, the unit establishes the hydraulic connections between the various engineering units of the water transfer system. With the goal of minimizing the total operating cost of the pumping stations in the water transfer system, the unit constructs a ten-day-scale optimized scheduling model for the inter-basin water transfer system. The actual water level of the lakes provides deterministic water level calculation boundaries for each section. Through the selection and switching of water sources and routes, the ten-day-scale scheduling scheme is determined.

[0030] The evaluation criteria establishment unit is used to calculate the tolerable unit water transfer energy consumption threshold based on the ten-day optimization scheduling model. When the actual unit water transfer energy consumption exceeds the threshold, it is considered that a unit water transfer energy consumption risk has occurred. At the same time, the risk of water transfer task is also considered to establish an evaluation criterion for the occurrence of water transfer risk.

[0031] The decision model construction unit is used to improve the ten-day optimal scheduling model based on the judgment criteria for the occurrence of water transfer risks, taking into account the uncertainty of water inflow, and using the water source inflow sequence of lakes along the water transfer route and the simulated values ​​of water level sequences of each lake along the water transfer route generated by the ten-day water level stochastic simulation model to provide the water level calculation boundary, and to construct a cross-basin scheduling risk quantification and optimization decision model.

[0032] The evaluation unit is used to obtain risk decision-making scheduling schemes for cross-basin water transfer systems based on cross-basin scheduling risk quantification and optimization decision-making models, and to evaluate the risks of the scheduling schemes and the sensitivity of key risk factors.

[0033] An electronic device for storing and executing the method, the device comprising:

[0034] Memory containing executable program code;

[0035] A processor coupled to the memory;

[0036] The processor calls the executable program code stored in the memory to execute the steps of the risk decision-making method for water allocation in inter-basin water transfer projects.

[0037] A computer-readable storage medium for storing and executing the method, the computer-readable storage medium storing computer instructions, which, when invoked, are used to execute the steps of the method for risk decision-making based on water allocation for inter-basin water transfer projects.

[0038] Beneficial effects: Compared with the prior art, the significant technical effects of the present invention are as follows: (1) It takes into account the uncertainty of water inflow in the water source area, assesses the water transfer risk rate by the completion of water transfer task and actual unit water transfer energy consumption, and constructs a scheduling optimization decision model that comprehensively considers cross-basin water volume scheduling optimization and risk value, which can provide low-risk, optimized cross-basin water volume optimization scheduling schemes and be applied to practical operation guidance; (2) Based on the cross-basin scheduling scheme risk and key risk factor sensitivity assessment method, it analyzes the impact of key risk factors in the water transfer process, and proposes optimization scheduling scheme preferences under different water inflow and operating conditions. Attached Figure Description

[0039] Figure 1 This is a flowchart of the method of the present invention;

[0040] Figure 2 This is a simplified schematic diagram of the ten-day water transfer system of the Jiangsu section of the South-to-North Water Diversion Project (Eastern Route) in an embodiment of the present invention.

[0041] Figure 3 This is a risk level classification diagram of unit water diversion energy consumption in an embodiment of the present invention;

[0042] Figure 4 The actual scheduling scheme for the Yangtze River and Huai River in 2018-2019 in the embodiments of the present invention is shown.

[0043] Figure 5 The optimized scheduling scheme for the Yangtze River and Huai River in 2018-2019, as described in the embodiments of the present invention, illustrates the utilization process of these water resources.

[0044] Figure 6 This is an example of the variation process of water transfer risk rate under different water transfer schemes in 2018-2019 in an embodiment of the present invention. Detailed Implementation

[0045] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0046] like Figure 1 As shown, the water allocation risk decision-making method for inter-basin water transfer projects according to the present invention specifically includes the following steps:

[0047] (1) Construct a ten-day optimal scheduling model for the cross-basin water transfer system and generate ten-day scheduling schemes based on the ten-day scale.

[0048] (1.1) Based on the objectives and influencing conditions of water diversion, the engineering relationships of lakes, rivers, pumping stations and gates in the water diversion system are spatially generalized, and the lakes are used as nodes to divide the system into sections.

[0049] (1.2) Based on the scheduling and operation rules and patterns of lakes, rivers, pumping stations, and sluice gates, scheduling simulation models for each engineering unit are established, including the following aspects:

[0050] (1.2.1) Based on the hydraulic characteristics and scheduling rules of the regulating lake, the operation mode is simulated, the regulating lake unit is simulated, and a lake scheduling simulation model is established.

[0051] (1.2.2) Taking water balance as the basic equation, considering the water exchange between the river channel and the upstream and downstream pumping stations as well as the water conveyance loss of the river section, the river channel unit is simulated to establish a river channel scheduling simulation model.

[0052] (1.2.3) Establish a pump station scheduling simulation model:

[0053] (1.2.3.1) The pumping volume of each pumping station during a certain period is calculated based on its average pumping flow rate during that period, using the following formula:

[0054]

[0055] in, For number i n The average pumping flow rate of the pumping station during that period; For number i n The pump station's total pumping volume in time period t; Δt is the length of the time period.

[0056] (1.2.3.2) Pump station operating cost: The power consumption of each pump station is used as the operating cost of each pump station, and the formula is as follows:

[0057]

[0058] in, For number i n The power consumption of the pump station during time period t; For number i n The total pumping volume of the pumping station in time period t; For number i n The pumping station has an average pumping flow rate of Energy consumption per unit of pump station.

[0059] (1.2.4) Simulate the operation mode of the gate based on the gate scheduling rules, perform gate unit simulation, and establish a gate scheduling simulation model.

[0060] (1.3) The actual water level of the lake provides a definite water level calculation boundary for each section, and the decision-making on the ten-day scheduling scheme is made by selecting and switching water sources and routes;

[0061] Under the premise of completing the total water transfer task within the scheduling period, the hydraulic connection between various engineering units of the water transfer system is established based on the scheduling simulation model of each engineering unit. With the goal of minimizing the total operating cost of the pumping stations in the water transfer system, an objective function and its constraints for a ten-day optimal scheduling model of the inter-basin water transfer system are constructed. The ten-day water transfer volume of each level of gate pumping station in each time period is used as the decision variable. Based on the constraints of the ten-day optimal scheduling model of the inter-basin water transfer system, the objective function of the ten-day optimal scheduling model is optimally solved. A ten-day scheduling scheme is generated, and the actual water transfer volume and unit water transfer energy consumption of the water transfer system in each time period are obtained. The actual unit water transfer energy consumption is equal to the total operating cost E of the pumping stations. year The ratio of the total water transfer volume to the total water transfer task. The objective function of the ten-day optimal scheduling model for the inter-basin water transfer system is:

[0062]

[0063] Among them, E year t represents the total operating cost of the pumping station; t represents the time period number, and T represents the total number of time periods; i n N0 represents the number of pump stations; For number i n The power consumption of the pump station during time period t.

[0064] (1.4) The constraints of the model include total water transfer volume constraint, lake water balance constraint, pump station working capacity constraint, lake storage capacity constraint, pumping control water level constraint, river water conveyance capacity constraint, river water level constraint, and river water balance constraint.

[0065] (1.5) The solution steps for the model are as follows:

[0066] (1.5.1) Divide the scheduling period into T time periods based on ten-day periods, and denot the time period number as t; initialize t = 1.

[0067] (1.5.2) Combining the total water transfer volume constraint in the model constraint conditions, a certain step size is set within a certain range, and the remaining water transfer volume at the end of time period t is discretized into N points. The discrete point number is denoted as n0, and n0 is initialized to 1.

[0068] (1.5.3) Combining the deterministic water level prediction sequence of lake C with the lake scheduling simulation model, the amount of water W that needs to be pumped into lake C at the n0 discrete point is calculated. rl (t,n0).

[0069] (1.5.4) Assume that the amount of water W pumped into lake C is... rl In the interval (t, n0), the proportion of water pumped into lake C via water diversion line A is a (0 ≤ a ≤ 1), and the proportion of water pumped into lake C via water diversion line B is 1 - a. If n line n water diversion linesline If the water volume passing through the first water diversion line is ≥3), then the proportion of water volume passing through the second water diversion line is a1 (0≤a1≤1), the proportion of water volume passing through the second water diversion line is a2(1-a1)(0≤a1,a2≤1), and so on, until the water volume passing through the nth water diversion line is ≥3. line The water volume of each water diversion line accounts for the following percentage: If the allocation ratio is within the constraints, a certain step size is set to allocate the water volume W to lake C. rl (t,n0) is discrete into M points, and the discrete points are numbered as m0.

[0070] (1.5.5) Combining the river channel scheduling simulation model and the pump station scheduling simulation model, the water pumping volume of other pump stations along the water diversion route is calculated in reverse along the water diversion route, as well as the outflow volume W of lake D (the upstream lake of lake C along the water diversion route) pumped out at the m0 discrete point. ch (t,n0,m0).

[0071] (1.5.6) Combining the deterministic water level prediction sequence of lake D and water source H river Based on the water inflow prediction sequence and lake scheduling simulation model, the amount of water W that needs to be pumped into lake D at the discrete point m0 ​​is calculated. rh (t,n0,m0) and H river Water utilization.

[0072] (1.5.7) Assume that the amount of water W pumped into lake D is... rh In the equation (t, n0, m0), the proportion of water pumped into lake D via route A is b (0 ≤ b ≤ 1), and the proportion of water pumped into lake D via route B is 1 - b. If n line n water diversion lines line If the water volume passing through the first water diversion line is ≥3, then the proportion of water volume passing through the second water diversion line is b1 (0≤b1≤1), the proportion of water volume passing through the second water diversion line is b2 (1-b1) (0≤b1,b2≤1), and so on, until the water volume passing through the nth water diversion line is ≥3. line The water volume of each water diversion line accounts for the following percentage: If the allocation ratio is within the constraints, a certain step size is set to allocate the water volume W to lake D. rh (t,n0,m0) is discrete into P points, and the discrete point is numbered p.

[0073] (1.5.8) Combining the river channel scheduling simulation model and the scheduling pump station simulation model, the water extraction volume of other pump stations along the water diversion route is calculated in reverse along the water diversion route, and the outflow volume W of the river channel E (the upstream river channel of lake D along the water diversion route) needs to be extracted from the river channel at the p-th discrete point. river1 (t,n0,m0,p).

[0074] (1.5.9) Based on the river channel scheduling simulation model of river channel E, calculate the amount of water W that needs to be pumped into river channel E at the p-th discrete point. river2 (t,n0,m0,p).

[0075] (1.5.10) Assume that the amount of water W pumped into river channel E is... river2 In the equation (t, n0, m0, p), the proportion of water pumped into river E via line A is c (0 ≤ c ≤ 1), and the proportion of water pumped into river E via line B is 1 - c. If n line n water diversion lines line If the water volume passing through the first water diversion line is ≥3, then the proportion of water volume passing through the second water diversion line is c1 (0≤c1≤1), the proportion of water volume passing through the second water diversion line is c2(1-c1)(0≤c1,c2≤1), and so on, until the water volume passing through the nth water diversion line is ≥3. line The water volume of each water diversion line accounts for the following percentage: If the allocation ratio is within the constraints, a certain step size is set to allocate the water volume W to river channel E. river2 (t,n0,m0,p) is discretized into R points.

[0076] (1.5.11) Combining the river channel scheduling simulation model and the pump station scheduling simulation model, the water extraction volume of other pump stations along the water diversion route is calculated in reverse along the water diversion route, as well as the water volume from water source C. river The amount of water extracted and utilized.

[0077] (1.5.12) Calculate the system cost of the above n0×M×P×R discrete points, and store the minimum one.

[0078] (1.5.13) When there are multiple lakes upstream of lake E along the water diversion route, the same steps (1.5.3)-(1.5.6) above can be used to add sections to calculate the amount of water diverted from and into the lakes, the amount of water source utilization of the lake water, and the pumping volume of the pumping stations along the route.

[0079] (1.5.14) If n0 = N, then proceed to step (1.5.15); otherwise, let n0 = n0 + 1 and return to step (1.5.3).

[0080] (1.5.15) If t = T, then end the operation and output a set of ten-day scheduling schemes; otherwise, take the remaining water transfer task at the end of time period t as the remaining water transfer task at time period t+1, let t = t+1, and return to step (1.5.2).

[0081] (2) Calculate the tolerable unit water transfer energy consumption threshold based on the ten-day optimization scheduling model. When the actual unit water transfer energy consumption exceeds the threshold, it is considered that the unit water transfer energy consumption risk has occurred. At the same time, consider the risk of water transfer task and establish the judgment criteria for the occurrence of water transfer risk.

[0082] (2.1) Based on the Copula function, a stochastic simulation model of water level is established to simulate different water level sequences of the reservoir during the scheduling period. The simulated values ​​of these sequences, along with different total water transfer tasks, are input into the ten-day optimization scheduling model of the inter-basin water transfer system. Multiple scheduling processes are simulated. Using the dynamic programming method, according to the model solution steps described in step (1.5), the deterministic water level prediction sequences in (1.5.3) and (1.5.6) are changed to simulated values ​​of different water level sequences for solution. The decision is made on the ten-day water transfer volume of the main rivers and pumping stations during the scheduling period, generating multiple ten-day scheduling schemes and their total unit water transfer energy consumption simulation values.

[0083] (2.1.1) The algorithm for the stochastic simulation model of ten-day water level based on the Copula function includes the following steps:

[0084] (2.1.2) The measured lake water level data sequence {x t The marginal distribution function F is obtained by fitting the water level values ​​at different time periods. t (x t ), where x t Let be the lake water level at time t; based on the measured lake water level data and the lake water levels x in two adjacent time periods. t and x t-1 marginal distribution function value z t and z t-1 Fit the joint distribution function of the Copula for two adjacent time periods, and obtain x from equation (4) based on the selected Copula function. t conditional distribution function According to equation (5), the uniformly distributed random number ε t and z t-1 Substitution The result is z t Thus, x is derived. t The value of z is given; the simulated values ​​of the lake water level sequence are calculated sequentially. When solving, z is given. t The initial value z0∈(0,1) is used. Substituting the randomly generated uniformly distributed random number ε1 into the first equation of (5) yields the water level edge distribution function value z1 for the first time period. Substituting z1 into the second equation of (5) yields the simulated lake water level value x1 for the first time period. Similarly, substituting z1 and the randomly generated uniformly distributed random number ε2 into equation (5) yields the water level edge distribution function value z2 and the simulated lake water level value x2 for the second time period. The edge distribution function values ​​z3, z4, ..., z2 for subsequent time periods can be calculated sequentially. m and their corresponding simulated water level values ​​x3, x4, ..., x m .

[0085] (2.1.3) Water level values ​​x for each time period t The conditional distribution function is:

[0086]

[0087] Among them, z t and z t-1 x t and x t-1 The marginal distribution function value, i.e., F t (x t ) and F t-1 (x t-1 ); For the selected Copula function.

[0088] (2.1.4) The formula for calculating the marginal distribution function and the simulated water level is as follows:

[0089]

[0090] Where t is the time period number, t = 1, 2, ..., m length m length ε is the length of the lake water level sequence; t is a randomly generated random number that follows a uniform distribution in the (0,1) region during the t-th time period.

[0091] (2.2) The main steps to determine whether the risk of the water transfer mission has occurred are as follows:

[0092] (2.2.1) Determine the maximum water conveyance capacity of the river channel and pumping station based on the engineering conditions;

[0093] (2.2.2) Determine the ten-day operation time of each section of the water transfer system according to the ten-day scheduling plan; determine the actual water transfer volume of each section in different ten-day scheduling plans by combining the operation time and the maximum water transfer capacity;

[0094] (2.2.3) In different ten-day scheduling schemes, when the actual water transfer volume during a period is less than the water transfer task volume during that period, it is considered that a water transfer task risk has occurred.

[0095] (2.3) The main steps to determine whether the risk of unit water transfer energy consumption has occurred are as follows:

[0096] (2.3.1) Using Fisher's optimal segmentation method, the simulated values ​​of different total unit water transfer energy consumption generated in step (2.1) are divided into acceptable, tolerable, and unacceptable zones. The upper and lower limits of unit water transfer energy consumption for each zone are calculated (see Appendix). Figure 3 The upper limit of the tolerable zone is the threshold of tolerable unit water transfer energy consumption;

[0097] (2.3.2) Fisher's optimal segmentation method includes the following steps:

[0098] (2.3.3) Suppose there are n samples arranged in a certain order forming a sequence {x}.n}, each sample is measured for m indicators, and while maintaining the original order of the samples, the sample sequence {x} is... n} is divided into k classes: Where i1, i2, ..., i k For dividing points, their indices satisfy 1 = i1 <i2<…<i k ≤i k+1 -1 = n; The intra-segment class diameter is used to measure the degree of difference between samples within a class. Let class P = {x} i ,x i+1 ,x i+2 ,…,x j (i, j are sample numbers, j>i), the intra-segment diameter of this class is denoted as D(i,j), and the mean is _____. The smaller the class diameter within a segment, the smaller the difference between samples within the class. Calculate the classification loss function e[P(n,k)], and find a set of split points that minimize the value of e[P(n,k)]. The corresponding split is the optimal split. The rationality of the optimal split is verified by the F-test.

[0099] (2.3.4) The relation matrix X can be constructed based on the sample feature values ​​as follows:

[0100]

[0101] Where, x nm This represents the value of the m-th indicator in the n-th sample.

[0102] (2.3.5) The formulas for calculating the class diameter and the mean eigenvalue are as follows:

[0103]

[0104] Where D(i,j) is the class {x i ,x i+1 ,x i+2 ,…,x j The intra-segment diameter of}; Let be the sample feature value of the r0th sample (i≤r0≤j); This represents the mean of the feature values ​​from the i-th sample to the j-th sample.

[0105] (2.3.6) The formula for calculating the classification loss function is:

[0106]

[0107] Where D(i) r i r+1 -1) is a class The segment diameter is r; r is the class number (r = 1 to k).

[0108] (2.3.7) The recursive formula for the optimal partition is:

[0109]

[0110] Where e[P(i-1,k-1)] is the loss function corresponding to the optimal segmentation that divides i-1 samples into k-1 classes; D(i,n) is the class {x i ,x i+1 ,x i+2 ,…,x n The diameter of the segment within}.

[0111] (2.3.8) The F-test formula is:

[0112]

[0113] Where, n r Let be the number of samples in the r-th class of sample sequences; Let be the sample mean of the r-th class of sample sequences; The mean of all samples; is the inter-segment class diameter; s is the sample number in the r-th class sample sequence; n rs Let be the value of the s-th sample in the r-th sample sequence.

[0114] (2.3.9) In different ten-day scheduling schemes, when the actual unit water transfer energy consumption is greater than the tolerable unit water transfer energy consumption threshold, it is considered that a unit water transfer energy consumption risk has occurred.

[0115] (2.4) The occurrence of either the water transfer task risk or the unit water transfer energy consumption risk is considered a water transfer risk. The frequency of water transfer risk occurrence under different ten-day scheduling schemes is statistically analyzed as a probability value to derive the water transfer risk rate, calculated using the following formula:

[0116]

[0117] in, For the kth water diversion system i Water transfer risk rate at each stage; P(n) k W num W task C num C threshold ) represents the probability function of the occurrence of water diversion risk; num represents the group number of the simulated lake water level sequence used for the water level calculation boundary; W num W represents the actual water diversion volume when the simulated water level sequence of the num-th lake is used as the boundary for water level calculation; task C represents the total water transfer task. numC represents the actual unit energy consumption for water diversion when the simulated water level sequence of the num-th lake is used as the boundary for water level calculation; threshold The threshold for tolerable unit water transfer energy consumption.

[0118] (3) Based on the established criteria for judging the occurrence of water transfer risks, the ten-day optimization scheduling model is improved, and then a cross-basin scheduling risk quantification and optimization decision-making model is constructed.

[0119] Based on the ten-day optimal scheduling model of the inter-basin water transfer system, considering the uncertainty of inflow, the water level calculation boundary is provided by the inflow sequence of the lakes along the water transfer route and the simulated values ​​of the water level sequence of each lake along the water transfer route generated by the ten-day random water level simulation model in step (2.1). A cross-basin scheduling risk quantification and optimization decision-making model is then constructed. By measuring the water transfer risk rate caused by uncertainty and coordinating the contradiction between economic benefits and risks, the water transfer cost is effectively reduced while ensuring the safe operation of the water transfer system and completing the total water transfer task. The main steps are as follows:

[0120] (3.1) Based on the objectives and influencing conditions of water transfer, the water transfer system is spatially generalized and divided into sections, as in step (1.1);

[0121] (3.2) Establish scheduling simulation models for each engineering unit based on the scheduling and operation rules and patterns of lakes, rivers, pumping stations and sluice gates, in the same manner as step (1.2);

[0122] (3.3) Considering the uncertainty of water inflow, the scheduling period is divided into K0 stages, based on the water inflow sequence of lakes along the water diversion route and the n generated by the stochastic simulation model of the water level in step (2.1) for ten days. k The simulated water level sequence of each lake along the water diversion route provides the water level calculation boundary for each section. The objective function is to minimize the water diversion risk rate of the water diversion system in each stage. The constraints are the same as in (1.4). A cross-basin scheduling risk quantification and optimization decision-making model is constructed. Through the selection and switching of water sources and routes, the risk decision-making scheduling scheme of the cross-basin water diversion system at the ten-day scale is determined, including the ten-day water pumping volume of each level of gate pumping station, the water diversion risk rate and benefit calculation within the time period.

[0123] (3.4) When solving the cross-basin scheduling risk quantification and optimization decision-making model, the scheduling period is divided into K0 stages. A reverse decision-making process is adopted, and the scheduling scenarios of each stage are iteratively simulated to solve the model. The last stage of the scheduling period (stage K0) is taken as the first stage of the decision-making process. The n generated by the ten-day water level stochastic simulation model is used. k The simulated water level sequence of the lakes is used to calculate the water transfer risk rate corresponding to each simulated water level sequence. The water transfer task volume with the lowest water transfer risk rate is selected as the stage water transfer volume. The calculation is repeated for the previous stage to obtain the water transfer volume, corresponding water transfer risk rate and benefits for each stage.

[0124] In this embodiment, the last stage of the scheduling period, K0 (late March to late May), is taken as the first stage for calculation, and the calculation proceeds sequentially backward. The specific steps are as follows:

[0125] (3.4.1) Determine the k-th i Water transfer task volume W during the phase (late March to late May) task,4 (where W) min <W task,4 <W max W min To meet the minimum water diversion requirements, W max To maximize the water diversion volume (affected by the pumping station's water diversion capacity), n is first generated using a stochastic simulation model of the ten-day water level. k Group k i The simulated water level sequence of the lake in each stage is calculated first using the water regulation risk rate quantification method described in step (2.4). i The stage water transfer risk rate corresponding to the simulated water level sequence for each stage is selected as the stage water transfer task amount with the lowest stage water transfer risk rate as W for this stage. task,4 .

[0126] (3.4.2) Determine the water transfer task W for stage k0-1 (late March to late May). task,3 Generate n k Group (k) i -1)-k i Simulated values ​​of lake water level sequence for each stage, and calculation of W task,3 Risk rate of staged water transfer under simulated water level sequences for each group.

[0127] (3.4.3) Determine (k) i -2) stage to (k i -1) Determine the minimum water transfer risk rate for the first stage, and so on, using the same method as step (3.4.2).

[0128] (3.4.4) Calculate the total water diversion risk rate (Risk) of the risk decision-making and scheduling scheme. w The calculation formula is as follows:

[0129]

[0130] (4) Assess the risk rate and sensitivity of key risk factors of cross-basin risk decision-making and scheduling schemes.

[0131] (4.1) Set different total water transfer tasks, water transfer start times, and first-stage water transfer volumes as boundary conditions. Based on the cross-basin scheduling risk quantification and optimization decision-making model, derive the optimal solution, generating water transfer costs and risk rates for different risk decision-making scheduling schemes. Consider water transfer costs as benefits, and obtain the process of changes in water transfer risk rate and benefits. Analyze the impact of different risk decision-making scheduling schemes on the relationship between water transfer risk rate and benefits from three perspectives: water transfer start time, first-stage water transfer volume, and main water source utilization. This provides a basis for weighing the allocation of water transfer volumes in different time periods during the operation of the cross-basin water transfer system. The specific steps are as follows:

[0132] (4.1.1) Calculate the water transfer risk rate by time period. Divide the scheduling period into multiple time periods. Assume that the water transfer starts in the first time period. Based on the total water transfer task, input the randomly generated n from the simulation. k After simulating the lake water level sequence, calculate the water diversion risk rate at this point; update the water diversion task volume for the next time period, which is equal to the total water diversion task volume minus the total water diversion volume before this time period, and then input the randomly generated n from the simulation. k The risk rate of water diversion at each lake level sequence is calculated, and the risk for each subsequent time period is calculated accordingly.

[0133] (4.1.2) Using different total water transfer tasks as boundary conditions, analyze the impact of water transfer start time on water transfer risk rate when the start time is in different time periods;

[0134] (4.1.3) Using different first-stage water transfer volumes as boundary conditions, analyze and compare the water transfer risk rates corresponding to different water transfer start times, and analyze the impact of the first-stage water transfer volume on the water transfer risk rate.

[0135] (4.1.4) During the scheduling process, due to the existence of multiple water sources, the utilization of the main water sources in different scheduling schemes varies under the same total water transfer task. Based on the different utilization of the main water sources, the impact of different water source utilization on the water transfer risk rate is compared and analyzed.

[0136] (4.2) Based on historical scheduling data, under a given total water transfer task, compare and analyze the actual historical scheduling schemes with the cross-basin scheduling risk quantification and optimization decision-making model to obtain the scheduling risk rate and benefit change process of the risk decision scheduling scheme, and analyze the impact of the previous decision on the subsequent process.

[0137] The water transfer risk rate of the historical actual scheduling scheme is calculated by generating a simulation scheme based on the actual water transfer volume process of the project every ten days and the simulated values ​​of different lake water level sequences generated by the Copula function. The actual water transfer volume and total water transfer task volume in the historical actual scheduling scheme are constant values, and the tolerable unit water transfer energy consumption threshold is consistent with that in the risk decision scheduling scheme. The water transfer risk rate of the historical actual scheduling scheme is calculated by statistical calculation of different actual unit water transfer energy consumption under different simulated values ​​of different lake water level sequences using equations (12) and (13).

[0138] (4.3) The sensitivity assessment method for key risk factors is as follows: First, a total water transfer task and simulated values ​​of lake water level sequences along the route are selected as benchmark conditions, and the risk rate of the benchmark scheme is obtained by the cross-basin scheduling risk quantification and optimization decision-making model. Then, the sensitivity coefficient of each key risk factor is obtained based on the change in water transfer risk rate caused by the change of each key risk factor.

[0139] (4.3.1) Key risk factors include the total amount of water diversion, the abundance or scarcity of water inflow to lakes, and the control level of the reservoirs.

[0140] (4.3.2) The formula for calculating the sensitivity coefficient is:

[0141]

[0142] Where S is the sensitivity coefficient, ΔRisk w ΔF represents the percentage change in risk, where ΔF is the percentage change in a key risk factor.

[0143] Example verification:

[0144] The embodiments of the present invention are specifically illustrated through the scheduling of the Jiangsu section of the first phase of the South-to-North Water Diversion Project (Eastern Route).

[0145] The Jiangsu section of the first phase of the South-to-North Water Diversion Project's eastern route utilizes the Jiangsu Provincial Water Diversion Project, expanding its scale and extending northward. Specifically, the Jiangsu section of the eastern route, based on the existing Grand Canal as the main water conveyance line, has formed a dual-line water conveyance pattern within Jiangsu through the construction, expansion, and reinforcement of pumping stations. The project uses the Yangtze and Huai Rivers as its main water sources, and includes regulating lakes such as Hongze Lake and Luoma Lake along the route. The Jiangsu section of the eastern route spans multiple river basins, including the Jianghuai and Yi-Shu-Si Rivers, and is a massive inter-basin water transfer project benefiting over 100 million people. It is responsible for conveying water to Shandong Province and meeting the water needs of the receiving areas within Jiangsu Province along the route. The specific steps for the scheduling risk identification and optimization decision-making methods of the Jiangsu section of the first phase of the South-to-North Water Diversion Project's eastern route are as follows:

[0146] (1) Establish a ten-day optimized scheduling model for the Jiangsu section of the first phase of the South-to-North Water Diversion Project (Eastern Route), and generate scheduling schemes based on ten-day periods, such as... Figure 2This paper generalizes the engineering relationships of lakes, rivers, pumping stations, and sluice gates within the Jiangsu section of the first phase of the South-to-North Water Diversion Project (Eastern Route). Using Hongze Lake and Luoma Lake as nodes, the water diversion project is divided into three sections: the Yangtze River-Hongze Lake section, the Hongze Lake-Luoma Lake section, and the section from Luoma Lake to the provincial exit. Each section includes two main water conveyance lines: the western section and the canal line. Each section's water conveyance line is further divided into several waterways by three cascade pumping stations. Scheduling simulation models are established for each engineering unit, based on the rules and patterns of scheduling and operation of lakes, rivers, pumping stations, and sluice gates. The objective function of the model is to minimize the operating cost of the pumping stations. The model constraints include the total water diversion task, lake water balance, pumping station capacity, lake storage capacity, pumping control water level, river water conveyance capacity, river water level, and river water balance. Using the real-time water levels of Hongze Lake and Luoma Lake as the boundary for water level calculation in each section, the system selects and switches water sources and routes, makes decisions on the scheduling scheme at the ten-day scale, and obtains the water transfer volume and actual unit water transfer energy consumption for each time period.

[0147] (2) Calculate the tolerable unit water transfer energy consumption threshold based on the ten-day optimization scheduling model. When the actual unit water transfer energy consumption exceeds the threshold, it is considered that the unit water transfer energy consumption risk has occurred. At the same time, consider the risk of water transfer task and establish the judgment criteria for the occurrence of water transfer risk.

[0148] Using ten-day average water level data from 2000 to 2020 (20 years in total), a stochastic simulation model of ten-day water level was established based on the Copula function to simulate the ten-day water level sequences of Hongze Lake and Luoma Lake. The RMSE values ​​of the mean, variance, and coefficient of variation were compared between historical and simulated water levels. If the water level sequence simulation performs well, it can be used as the water level input for cross-basin scheduling risk quantification and optimization decision-making models.

[0149] Simulated values ​​of different water level sequences for Hongze Lake and Luoma Lake, as well as different total water diversion task volumes (W) task =703 million, 844 million, 884 million m 3 Input the inter-basin water transfer system's ten-day optimal scheduling model and run the simulation to obtain 300 sets of ten-day scheduling schemes.

[0150] Based on the engineering conditions, determine the maximum water conveyance capacity of the river channel and pumping stations; determine the ten-day operating time of each section of the water transfer system according to the 300-group ten-day scheduling plan; and determine the actual water transfer volume for each section in the 300-group ten-day scheduling plan by combining the operating time and the maximum water conveyance capacity. If the actual water transfer volume is less than the water transfer task volume for the time period, it is considered that a water transfer task risk has occurred.

[0151] Frequency analysis was performed on 300 sets of scheduling results. Using Fisher's optimal partitioning method, the simulated energy consumption per unit of water transfer was divided into acceptable, tolerable, and unacceptable zones. The upper and lower limits of energy consumption per unit of water transfer for each zone were then determined: Acceptable zone (0.0467 kWh / m³). 3 0.0603 kWh / m3 Tolerable range (0.0603 kWh / m) 3 0.1015 kWh / m 3 ) and unacceptable zone [0.1015kWh / m 3 0.1126 kWh / m 3 When the actual unit water transfer energy consumption is greater than 0.1015 kWh / m³ 3 When this occurs, it is considered a risk of energy consumption per unit of water diversion.

[0152] (3) Constructing a quantification and optimization decision-making model for scheduling risks in the Jiangsu section of the first phase of the South-to-North Water Diversion Project (Eastern Route). Based on the objectives and influencing conditions of water diversion, the Jiangsu section of the South-to-North Water Diversion Project (Eastern Route) is generalized and divided into sections. For the scheduling and operation rules and patterns of lakes, rivers, pumping stations, and sluice gates along the route, scheduling simulation models for each engineering unit are established, using the same method as the ten-day optimization scheduling model for the Jiangsu section of the first phase of the South-to-North Water Diversion Project (Eastern Route). Considering the uncertainty of inflow, the scheduling period is divided into K0 stages. The simulated values ​​of 300 sets of lake water level sequences are used to provide water level calculation boundaries for each section. The total water diversion task, lake water balance, pumping station capacity, lake storage capacity, pumping control water level, river water conveyance capacity, river water level, and river water balance are used as model constraints. A cross-basin scheduling risk quantification and optimization decision-making model is constructed to determine the ten-day-scale scheduling scheme of the system and calculate the water diversion risk rate and benefits.

[0153] The objective function of the cross-basin scheduling risk quantification and optimization decision-making model is the above formula (14).

[0154] (4) Assess the risks of cross-basin scheduling schemes, the main contents of which are as follows:

[0155] Different total water transfer volumes are used as boundary conditions. The Jiangsu section of the first phase of the South-to-North Water Transfer Project (Eastern Route) typically begins water transfer in early December. If water conditions are favorable in November, pre-transfer will be conducted in November. Therefore, the water transfer start time is divided into three stages: early to late November; early December to late January; and early February to early March. The impact of the water transfer start time at different stages on the water transfer risk rate is analyzed.

[0156] Using the water transfer volume in the first phase (early November to late November) as the initial water transfer volume, different initial water transfer volumes were set as boundary conditions. The water transfer risk rates corresponding to different start times were compared, and the impact of the initial water transfer volume on the water transfer risk rate was analyzed.

[0157] Using a specific total water transfer target as input, two scheduling schemes are set up. The water sources for the Jiangsu section of the first phase of the South-to-North Water Transfer Project (Eastern Route) include the Yangtze River and the Huai River (referred to as Yangtze River water and Huai River water, respectively). Therefore, Scheme 1 focuses on using Yangtze River water, while Scheme 2 focuses on using Huai River water. The impact of the utilization of Yangtze River and Huai River water on the water transfer risk rate is compared.

[0158] Taking the water transfer in 2018-2019 as an example, with a total water transfer volume of 844 million cubic meters... 3 Using these as input conditions, the risk quantification and optimization decision-making model for cross-basin scheduling is used to obtain risk-determined scheduling schemes (see attached). Figure 4 ) and the actual scheduling scheme (attached) Figure 5 A comparison was made to analyze the dynamic change process of the water transfer risk rate (see appendix). Figure 6 ), and the impact of pre-transfer water transfer decisions on subsequent processes.

[0159] (5) Assess the sensitivity of key risk factors and propose preferred scheduling schemes under different inflow and operating conditions. Key risk factors are set as total water transfer volume, water levels in Hongze Lake and Luoma Lake (both wet and dry seasons), and the control water level of Luoma Lake. The parameter values ​​corresponding to the baseline scheme for sensitivity analysis are determined: Total water transfer volume W task =884 million m 3 The baseline for lake inflow abundance / scarcity is based on statistical analysis from 2000 to 2020, with the control water level of Luoma Lake set at 23.0 m. The water transfer risk rate corresponding to the baseline scheme is calculated, with the first phase starting in early November, the second phase in early December, and the third phase in early March. Given the total water transfer workload, the percentage changes in four key risk factors—abundance / scarcity of inflow to Hongze Lake, abundance / scarcity of inflow to Luoma Lake, and the control water level of Luoma Lake—the water transfer risk rate and sensitivity coefficient under different scheduling conditions caused by changes in these key risk factors are calculated. Based on this, a set of typical scheduling scenarios and operating conditions is established, and preferred scheduling schemes are proposed under different inflow and operating conditions.

[0160] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A risk decision-making method for water allocation in inter-basin water transfer projects, characterized in that, Includes the following steps: Spatially generalize the relationships between various engineering units within the inter-basin water transfer system, dividing the system into sections with lakes as nodes; and establish scheduling simulation models for each engineering unit. Based on these simulation models, establish hydraulic connections between the engineering units of the water transfer system. With the goal of minimizing the total operating cost of the pumping stations in the water transfer system, construct a ten-day optimal scheduling model for the inter-basin water transfer system. Use the actual water level of the lakes as the deterministic water level calculation boundary for each section, and make decisions on the ten-day scheduling scheme through the selection and switching of water sources and routes. The tolerable unit water transfer energy consumption threshold is calculated based on the ten-day optimized scheduling model. When the unit water transfer energy consumption exceeds this threshold, it is considered that a unit water transfer energy consumption risk has occurred. Simultaneously considering the risk of the water transfer task, a criterion for judging the occurrence of water transfer risk is established. The criterion for judging the occurrence of water transfer risk is as follows: A stochastic simulation model of the water level of lakes along the water transfer route is established based on the Copula function. The simulated values ​​of the lake water level sequence for each ten-day period are simulated. The simulated values ​​are then input into the ten-day optimal scheduling model of the inter-basin water transfer system along with different total water transfer task amounts. Multiple scheduling processes are simulated, and multiple ten-day scheduling schemes and their total unit water transfer energy consumption simulation values ​​are generated. Determine the maximum water conveyance capacity of the river channel and pumping station, and determine the ten-day operation time of each section of the water transfer system according to the ten-day scheduling plan; combine the operation time and the maximum water conveyance capacity to determine the actual water transfer volume of each section in different ten-day scheduling plans; Fisher's optimal segmentation method was used to divide the simulated values ​​of different total unit water transfer energy consumption into acceptable, tolerable and unacceptable zones. The upper and lower limits of unit water transfer energy consumption in each zone were calculated, and the upper limit of the tolerable zone was the tolerable unit water transfer energy consumption threshold. When the actual water transfer volume during a given period is less than the water transfer task volume during that period, it is considered that a water transfer task risk has occurred; when the actual unit water transfer energy consumption is greater than the tolerable unit water transfer energy consumption threshold, it is considered that a unit water transfer energy consumption risk has occurred; when either the water transfer task risk or the unit water transfer energy consumption risk occurs, it is considered that a water transfer risk has occurred. Based on the criteria for assessing the occurrence of water transfer risks, the ten-day optimization scheduling model is improved. Taking into account the uncertainty of water inflow, the simulated values ​​of water level sequences of each lake along the water transfer route generated by the ten-day water level stochastic simulation model are used to provide the water level calculation boundary, and a cross-basin scheduling risk quantification and optimization decision-making model is constructed. Based on the risk quantification and optimization decision-making model of cross-basin scheduling, a risk decision-making scheduling scheme for cross-basin water transfer system is obtained, and the risk of the scheduling scheme and the sensitivity of key risk factors are evaluated.

2. The risk decision-making method for water allocation in inter-basin water transfer projects according to claim 1, characterized in that, The objective function of the ten-day optimal scheduling model for inter-basin water transfer systems is: ； in, This represents the total operating cost of the pumping station. Here, T represents the time period number, and T represents the total number of time periods. Number the pump station Number of pumping stations; For the number pump station Power consumption during a given time period; The constraints of the model include: total water diversion volume constraint, lake water balance constraint, pumping station working capacity constraint, lake storage capacity constraint, pumping control water level constraint, river water conveyance capacity constraint, river water level constraint, and river water balance constraint.

3. The risk decision-making method for water allocation in inter-basin water transfer projects according to claim 1, characterized in that, The method for constructing a risk quantification and optimization decision-making model for inter-basin water transfer is as follows: Based on the objectives and influencing conditions of water transfer, the inter-basin water transfer system is spatially generalized and divided into sections; for the scheduling and operation rules and patterns of lakes, rivers, pumping stations, and sluice gates, scheduling simulation models for each engineering unit are established; considering the uncertainty of inflow, the scheduling period is divided into... Each stage is generated using a stochastic simulation model of the water level over ten days. The group of water source water levels provides water level calculation boundaries for each section. With the objective function of minimizing the stage water transfer risk rate of the water transfer system, a cross-basin scheduling risk quantification and optimization decision-making model is constructed. The constraints of the model include: total water transfer volume constraint, lake water balance constraint, pump station working capacity constraint, lake storage capacity constraint, pumping control water level constraint, river water conveyance capacity constraint, river water level constraint, and river water balance constraint.

4. The risk decision-making method for water allocation in inter-basin water transfer projects according to claim 1, characterized in that, The solution method for the cross-basin scheduling risk quantification and optimization decision-making model is as follows: A reverse-order decision-making process is used to iteratively simulate the scheduling scenarios at each stage to solve the model, with the last stage within the scheduling period as the solution. As the first stage of decision-making, the water level was generated using a stochastic simulation model. The simulated water level series of lakes were used to calculate the water diversion risk rate corresponding to each set of simulated water level series. The water diversion task volume with the lowest water diversion risk rate was selected as the stage water diversion volume. The calculation was repeated for the previous stage to obtain the water diversion volume, corresponding water diversion risk rate, and benefits for each stage. The overall water diversion risk rate of the risk decision scheduling scheme was also calculated. The calculation formula is: ； ； in, For the water diversion system The risk rate of water diversion during each stage; A probability function representing the risk of water diversion occurring; Indicates the group number of the simulated lake water level sequence used for the water level calculation boundary; For the first The actual water diversion volume when using the simulated values ​​of the lake water level sequence as the boundary for water level calculation; This represents the total water transfer task. For the first The actual unit energy consumption for water diversion when using the simulated values ​​of the lake water level sequence as the boundary for water level calculation; The threshold for tolerable unit water transfer energy consumption.

5. The risk decision-making method for water allocation in inter-basin water transfer projects according to claim 1, characterized in that, The risk assessment method for the scheduling scheme is as follows: different total water transfer tasks, water transfer start times, and first-stage water transfer volumes are set as boundary conditions. The optimal solution is obtained based on the cross-basin scheduling risk quantification and optimization decision-making model. The water transfer costs and water transfer risk rates of different risk decision scheduling schemes are generated. The water transfer costs are considered as benefits, and the process of water transfer risk rate and benefit changes is obtained. From the perspectives of water transfer start time, first-stage water transfer volume, and main water source utilization of different risk decision scheduling schemes, the impact of these factors on the relationship between water transfer risk rate and benefits is analyzed, providing a basis for weighing the allocation of water transfer volume in different time periods during the operation of the cross-basin water transfer system. Based on historical scheduling data, under a given total water transfer task, the scheduling risk rate and benefit changes of the risk decision scheduling scheme are obtained by comparing and analyzing the actual historical scheduling schemes with the cross-basin scheduling risk quantification and optimization decision-making model, and the impact of the previous decision on the subsequent process is analyzed.

6. The risk decision-making method for water allocation in inter-basin water transfer projects according to claim 1, characterized in that, The method for assessing the sensitivity of key risk factors is as follows: First, a total water transfer task and simulated values ​​of lake water level sequences along the route are selected as benchmark conditions, and the risk rate of the benchmark scheme is obtained by the cross-basin scheduling risk quantification and optimization decision-making model; then, the sensitivity coefficient of each key risk factor is obtained based on the change in water transfer risk rate caused by the change of each key risk factor. Key risk factors include the total water diversion volume, the inflow and outflow of water to lakes, and the controlled water level of the regulating lakes; the formula for calculating the sensitivity coefficient is: ； Where S is the sensitivity coefficient. The percentage change in risk rate. This represents the percentage change in a key risk factor.

7. A risk decision-making system for water allocation in inter-basin water transfer projects, characterized in that, include: The ten-day-scale scheduling scheme generation unit is used to spatially generalize the relationships between various engineering units within the inter-basin water transfer system, dividing the system into sections with lakes as nodes; and establishing scheduling simulation models for each engineering unit. Based on the scheduling simulation models of each engineering unit, the unit establishes the hydraulic connections between the various engineering units of the water transfer system. With the goal of minimizing the total operating cost of the pumping stations in the water transfer system, the unit constructs a ten-day-scale optimized scheduling model for the inter-basin water transfer system. The actual water level of the lakes provides deterministic water level calculation boundaries for each section. Through the selection and switching of water sources and routes, the ten-day-scale scheduling scheme is determined. The evaluation criteria establishment unit is used to calculate the tolerable unit water transfer energy consumption threshold based on the ten-day optimized scheduling model. When the actual unit water transfer energy consumption exceeds this threshold, it is considered that a unit water transfer energy consumption risk has occurred. Simultaneously, considering the risk of the water transfer task, an evaluation criterion for the occurrence of water transfer risk is established. The evaluation criteria for the occurrence of water transfer risk are as follows: A stochastic simulation model of the water level of lakes along the water transfer route is established based on the Copula function. The simulated values ​​of the lake water level sequence for each ten-day period are simulated. The simulated values ​​are then input into the ten-day optimal scheduling model of the inter-basin water transfer system along with different total water transfer task amounts. Multiple scheduling processes are simulated, and multiple ten-day scheduling schemes and their total unit water transfer energy consumption simulation values ​​are generated. Determine the maximum water conveyance capacity of the river channel and pumping station, and determine the ten-day operation time of each section of the water transfer system according to the ten-day scheduling plan; combine the operation time and the maximum water conveyance capacity to determine the actual water transfer volume of each section in different ten-day scheduling plans; Fisher's optimal segmentation method was used to divide the simulated values ​​of different total unit water transfer energy consumption into acceptable, tolerable and unacceptable zones. The upper and lower limits of unit water transfer energy consumption in each zone were calculated, and the upper limit of the tolerable zone was the tolerable unit water transfer energy consumption threshold. When the actual water transfer volume during a given period is less than the water transfer task volume during that period, it is considered that a water transfer task risk has occurred; when the actual unit water transfer energy consumption is greater than the tolerable unit water transfer energy consumption threshold, it is considered that a unit water transfer energy consumption risk has occurred; when either the water transfer task risk or the unit water transfer energy consumption risk occurs, it is considered that a water transfer risk has occurred. The decision model construction unit is used to improve the ten-day optimal scheduling model based on the judgment criteria for the occurrence of water transfer risks, taking into account the uncertainty of water inflow, and using the water source inflow sequence of lakes along the water transfer route and the simulated values ​​of water level sequences of each lake along the water transfer route generated by the ten-day water level stochastic simulation model to provide the water level calculation boundary, and to construct a cross-basin scheduling risk quantification and optimization decision model. The evaluation unit is used to obtain risk decision-making scheduling schemes for cross-basin water transfer systems based on cross-basin scheduling risk quantification and optimization decision-making models, and to evaluate the risks of the scheduling schemes and the sensitivity of key risk factors.

8. An electronic device, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the steps of the water quantity scheduling risk decision-making method for inter-basin water transfer projects as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, which, when invoked, are used to execute the steps of the water quantity scheduling risk decision-making method for inter-basin water transfer projects as described in any one of claims 1-6.

Images