Drainage basin discharge capacity distribution optimization method based on data driving

Abstract

The invention relates to the technical field of hydraulic engineering, and discloses a drainage basin pollution discharge capacity distribution optimization method based on data driving, which comprises the following steps: obtaining drainage basin pollution discharge data; the method comprises the following steps: determining a water environment bearing index according to drainage basin pollution discharge data, and constructing a drainage basin pollution discharge capacity distribution double-layer optimization model by taking minimization of pollution treatment cost as an upper layer target and maximization of pollution discharge configuration income as a lower layer target; a water environment bearing index uncertainty set is introduced into the drainage basin discharge capacity distribution double-layer optimization model, and a drainage basin discharge capacity distribution double-layer robust optimization model is constructed; and converting the drainage basin discharge capacity distribution double-layer robust optimization model into a single-layer optimization model by adopting a KKT condition, and solving to obtain an optimal drainage basin discharge capacity distribution result. According to the method, the key index of the water environment bearing index is considered, more decision information and environmental factors are comprehensively considered in the model, and more accurate drainage basin discharge capacity decision distribution is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of water conservancy projects, and in particular to a data-driven method for optimizing watershed discharge allocation. Background Art

[0002] River basins are crucial carriers of the water environment. While supporting various economic and social activities along their shores, they also bear a significant burden of ecological and environmental pollution. The allocation of river basin emission rights is a multi-stakeholder system optimization problem, involving the balance between multiple objectives, making allocation decisions more complex. Therefore, in optimizing the allocation of river basin emission rights, how to coordinate ecological protection compensation, coordinate the conflicts of interest among multiple stakeholders, and achieve a scientific and efficient allocation of river basin emission rights is a complex and systemic challenge that urgently needs to be addressed.

[0003] In pollution control practices, traditional methods for allocating emission rights often rely on static indicators, ignoring the hydrological conditions of a river basin, the spatiotemporal heterogeneity of environmental capacity, and the migration and diffusion patterns of pollutants. This leads to a disconnect between total emission control and the region's actual carrying capacity. Furthermore, pollution source data is scattered across multiple, heterogeneous systems, including enterprise self-reporting, environmental monitoring, and satellite remote sensing. This data suffers from inconsistent coverage, delayed updates, and questionable authenticity, leading to decision-making relying on fragmented information. Overcoming core bottlenecks in data integration, dynamic modeling, and collaborative optimization to achieve the scientific quantification and precise regulation of environmental rights and interests has become a pressing issue in water environment governance. Summary of the Invention

[0004] In view of the above-mentioned deficiencies in the prior art, the present invention provides a data-driven method for optimizing watershed discharge allocation.

[0005] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is: A data-driven basin discharge allocation optimization method includes the following steps: Obtain basin pollution data; The water environment carrying index is determined based on the basin pollution data. A two-level optimization model for basin pollution allocation is constructed with minimizing pollution control costs as the upper level goal and maximizing pollution allocation benefits as the lower level goal. The uncertainty set of water environment carrying index is introduced into the two-layer optimization model of watershed discharge allocation to construct a two-layer robust optimization model of watershed discharge allocation. The KKT condition is used to transform the two-layer robust optimization model of basin discharge allocation into a single-layer optimization model, and the optimal basin discharge allocation result is obtained.

[0006] Furthermore, the water environment carrying index is determined based on the basin pollution data as follows:

[0007] in,x ik Assigned to the region i pollutants k The amount of sewage discharged, λ ik For the region i Pollutants k The governance ratio, M A collection of regions.

[0008] Furthermore, with minimizing pollution control costs as the upper-level goal and maximizing the benefits of pollution discharge allocation as the lower-level goal, a two-level optimization model for watershed pollution discharge allocation is constructed, including: Taking minimizing pollution control costs as the upper-level goal, and taking ecological compensation cost constraints, water environment carrying index constraints, discharge demand constraints, emission reduction rate constraints, and parameter non-negativity constraints as constraints, the upper-level model is constructed. The lower-level model is constructed with the maximization of pollution allocation benefits as the lower-level goal, and with the constraints that the pollutant emissions of each polluting enterprise do not exceed the pollutant allocation amount of the region to which it belongs, the pollutant emissions of each polluting enterprise do not exceed the minimum and maximum demand amounts, and the non-negative parameters. A two-layer optimization model for basin discharge allocation is constructed based on the upper model and the lower model.

[0009] Furthermore, with minimizing the pollution control cost as the upper-level goal, and with the ecological compensation cost constraint, water environment carrying index constraint, pollutant discharge demand constraint, emission reduction rate constraint, and parameter non-negative constraint as the constraint conditions, the upper-level model is constructed, specifically: ; st

[0010]

[0011]

[0012]

[0013]

[0014] in, F For pollution control costs, CF For ecological compensation costs, p jk For polluting enterprises j Acquisition of pollutants k The cost of emissions, y ijk For the region i Allocated to polluting enterprises j pollutantsk emissions, c ik For the region i Treating pollutants k the cost, x ik Assigned to the region i pollutants k emissions, K is a collection of pollutants, M is a set of regions, N For polluting enterprises, For the region i Contribution coefficient of participating in ecological compensation, For the region i The responsibility coefficient, is the ecological compensation accounting value, is the upstream region collection, is the downstream area collection, is the environmental carrying index of the watershed for pollutant k, For the region i pollutants k The lowest emissions, For the region i pollutants k Baseline emissions, For the region i pollutants k The minimum emission reduction rate, For the region i pollutants k the highest reduction rate.

[0015] Furthermore, with maximizing the benefits of pollution discharge allocation as the lower-level goal, and with the constraints that the pollutant emissions of each polluting enterprise do not exceed the pollutant allocation amount of the region to which they belong, the pollutant emissions of each polluting enterprise do not exceed the minimum and maximum requirements, and the parameters are non-negative, the lower-level model is constructed, specifically:

[0016] st

[0017] Among them, max is the maximum value function, f i For the region i The pollution discharge configuration income, For the region i Medium-sized polluting enterprises j pollutants k Emission benefits, y ijk For the region iAllocated to polluting enterprises j pollutants k emissions, p jk For polluting enterprises j Acquisition of pollutants k The cost of emissions, is the ecological compensation accounting value, x ik Assigned to the region i pollutants k emissions, K is a collection of pollutants, N For polluting enterprises, For the region i Medium-sized polluting enterprises j Emission of pollutants k The minimum demand, For the region i Medium-sized polluting enterprises j Emission of pollutants k The highest demand.

[0018] Furthermore, the uncertainty set of water environment carrying index is introduced into the two-layer optimization model of watershed discharge allocation as follows: The water environment carrying index constraint is transformed into uncertainty constraint, and the moment information fuzzy set of the water environment carrying index is used to represent the disturbance of uncertainty parameters. The moment information fuzzy set of water environment carrying index is reconstructed into a solvable event-type fuzzy set.

[0019] Furthermore, the disturbance of uncertainty parameters is expressed using the moment information fuzzy set of the water environment carrying index as follows:

[0020] in, is the moment information fuzzy set, is a random variable that obeys the joint probability distribution P, is a random variable The probability distribution family in K-dimensional real space, is a random variable under distribution P The mean of is the mean of the variance under distribution P, is the covariance mean under distribution P, is a random variable under distribution P Satisfy its upper and lower limits The mean of is the mean vector of random variables, is the mean water environment carrying index of pollutant k, is the covariance matrix of the random variables, is the standard deviation of the random variable, is a unit vector, T represents transpose, is the lower bound vector of the random variable distribution, is the upper bound vector of the random variable distribution, K A collection of pollutants.

[0021] Furthermore, the moment information fuzzy set of the water environment carrying index is reconstructed into a solvable event-type fuzzy set as follows:

[0022] in, is an event-type fuzzy set, For random events, is an auxiliary variable, representing a random variable and its mean The difference, is a random variable under the condition of the occurrence of type 1 random event The mean of Under the condition of the occurrence of type 1 random event Regarding the mean value of k for different pollutants, is a random variable under the condition of the occurrence of type 1 random event The covariance of Under the condition of the occurrence of type 1 random event and Belong to the set The mean probability of is a random variable under distribution P The mean that satisfies its lower and upper bounds.

[0023] Furthermore, a two-layer robust optimization model for basin discharge allocation is constructed as follows: .

[0024] Furthermore, the KKT condition is used to transform the two-layer robust optimization model of basin discharge allocation into a single-layer optimization model, and the optimal basin discharge allocation result is obtained, including: The KKT conditional method is used to transform the lower-level objective function and constraint conditions, and the original two-level optimization model is transformed into a single-level optimization model; Clarify the domain of the single-layer optimization model; Randomly select the initial optimal solution within the model domain and record it as ; Use gradient descent to find the optimal solution Iterate and finally get the optimal solution of the model ; Add fuzzy sets to obtain a series of optimal solutions of the optimization model under different uncertainty levels , to achieve the optimal allocation of river basin pollution discharge rights under uncertain environments.

[0025] The present invention has the following beneficial effects: The present invention considers the key indicator of water environment carrying index, and uses the method of distributed robustness to characterize the possible situations and problems. In the model, more decision-making information and environmental factors are comprehensively considered to achieve more accurate basin discharge decision-making and allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a flow chart of a data-driven basin discharge allocation optimization method. DETAILED DESCRIPTION

[0027] The specific embodiments of the present invention are described below to facilitate understanding of the present invention by those skilled in the art. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the appended claims, these changes are obvious, and all inventions and creations utilizing the concepts of the present invention are protected.

[0028] like Figure 1 As shown, the embodiment of the present invention provides a data-driven basin discharge allocation optimization method, including the following steps S1 to S4: S1. Obtain basin pollution data; S2. Determine the water environment carrying index based on the basin pollution data, and construct a two-level optimization model for basin pollution allocation with minimizing pollution control costs as the upper level goal and maximizing pollution allocation benefits as the lower level goal; In an optional embodiment of the present invention, step S2 determines the water environment carrying index based on the basin sewage discharge data as follows:

[0029] in, x ik Assigned to the region i pollutants k The amount of sewage discharged, λ ik For the region i Pollutants k The governance ratio, M A collection of regions.

[0030] This embodiment takes minimizing pollution control costs as the upper-level goal and maximizing the benefits of pollutant discharge allocation as the lower-level goal, and constructs a two-level optimization model for basin pollutant discharge allocation, including: Taking minimizing pollution control costs as the upper-level goal, and taking ecological compensation cost constraints, water environment carrying index constraints, discharge demand constraints, emission reduction rate constraints, and parameter non-negativity constraints as constraints, the upper-level model is constructed. The lower-level model is constructed with the maximization of pollution allocation benefits as the lower-level goal, and with the constraints that the pollutant emissions of each polluting enterprise do not exceed the pollutant allocation amount of the region to which it belongs, the pollutant emissions of each polluting enterprise do not exceed the minimum and maximum demand amounts, and the non-negative parameters. A two-layer optimization model for basin discharge allocation is constructed based on the upper model and the lower model.

[0031] This embodiment aims to minimize pollution control costs during the watershed pollution control process, and includes two parts: (1) pre-treatment of pollution that is about to be discharged into the water environment; (2) paid allocation of pollution discharge rights; and (3) encouraging the implementation of ecological compensation funds in lower-level regions. Therefore, the upper-level objective function is:

[0032] In the process of allocating pollutant emissions, it is necessary to clarify the accounting standards for ecological compensation funds and establish ecological compensation fund constraints:

[0033] When formulating the total amount of pollution discharge rights, the natural carrying capacity of the water environment must be taken into consideration, and pollution discharge must be kept within a reasonable range.

[0034] Different regions have different industrial structures, and the types and quantities of pollutants they emit vary. Therefore, the implementation of environmental policies needs to ensure the normal production and life of the economy and society. Therefore, pollution emission restrictions constrain:

[0035] Regions also need to conduct longitudinal comparisons over time and reduce emissions based on the previous year. However, there also needs to be an upper limit on emissions reductions to avoid excessive reductions that could affect normal production and life.

[0036] Parameter non-negativity constraint:

[0037] in, F For pollution control costs, CF For ecological compensation costs, p jk For polluting enterprises j Acquisition of pollutants k The cost of emissions,y ijk For the region i Allocated to polluting enterprises j pollutants k emissions, c ik For the region i Treating pollutants k the cost, x ik Assigned to the region i pollutants k emissions, K is a collection of pollutants, M is a set of regions, N For polluting enterprises, For the region i Contribution coefficient of participating in ecological compensation, For the region i The responsibility coefficient, is the ecological compensation accounting value, is the upstream region collection, is the downstream area collection, is the environmental carrying index of the watershed for pollutant k, For the region i pollutants k The lowest emissions, For the region i pollutants k Baseline emissions, For the region i pollutants k The minimum emission reduction rate, For the region i pollutants k the highest reduction rate.

[0038] This embodiment takes maximizing the benefits of pollutant emission allocation as the lower-level goal, and takes the following constraints: the pollutant emissions of each pollutant-emitting enterprise do not exceed the pollutant allocation amount of the region to which it belongs, the pollutant emissions of each pollutant-emitting enterprise do not exceed the minimum and maximum requirements, and the parameter non-negativity constraint, to construct the lower-level model:

[0039] st

[0040] Among them, max is the maximum value function, f i For the region i The pollution discharge configuration income, For the region i Medium-sized polluting enterprises jpollutants k Emission benefits, y ijk For the region i Allocated to polluting enterprises j pollutants k emissions, p jk For polluting enterprises j Acquisition of pollutants k The cost of emissions, is the ecological compensation accounting value, x ik Assigned to the region i pollutants k emissions, K is a collection of pollutants, N For polluting enterprises, For the region i Medium-sized polluting enterprises j Emission of pollutants k The minimum demand, For the region i Medium-sized polluting enterprises j Emission of pollutants k The highest demand.

[0041] S3. Introducing the uncertainty set of water environment carrying index into the two-layer optimization model of watershed discharge allocation, and constructing a two-layer robust optimization model of watershed discharge allocation; In an optional embodiment of the present invention, step S3 considers the uncertainty in the decision-making process and uses a robust optimization method to solve it, processing the uncertain set of environmental carrying capacity, including: S3-1, constrain Replace with To characterize uncertainty; S3-2. Collect moment information of water environment carrying capacity and use moment information fuzzy sets to represent the disturbance of random parameters:

[0042] in, is the moment information fuzzy set, is a random variable that obeys the joint probability distribution P, is a random variable The probability distribution family in K-dimensional real space, is a random variable under distribution P The mean of is the mean of the variance under distribution P, is the covariance mean under distribution P, is a random variable under distribution P Satisfy its upper and lower limits The mean of is the mean vector of random variables, is the mean water environment carrying index of pollutant k, is the covariance matrix of the random variables, is the standard deviation of the random variable, is a unit vector, T represents transpose, is the lower bound vector of the random variable distribution, is the upper bound vector of the random variable distribution, K A collection of pollutants.

[0043] S3-3. Reconstruct the fuzzy set of S3-2 into a solvable event-type fuzzy set form:

[0044] in, is an event-type fuzzy set, For random events, is an auxiliary variable, representing a random variable and its mean The difference, is a random variable under the condition of the occurrence of type 1 random event The mean of Under the condition of the occurrence of type 1 random event Regarding the mean value of k for different pollutants, is a random variable under the condition of the occurrence of type 1 random event The covariance of Under the condition of the occurrence of type 1 random event and Belong to the set The mean probability of is a random variable under distribution P The mean that satisfies its lower and upper bounds.

[0045] Its support set is:

[0046] in, Represents event-type fuzzy sets; random vectors and random scenarios They are all auxiliary variables.

[0047] S3-4. Through the above steps, a two-layer robust optimization allocation model for river basin pollution discharge rights is finally constructed:

[0048] in, .

[0049] S4. Use KKT conditions to transform the two-layer robust optimization model of watershed discharge allocation into a single-layer optimization model, and obtain the optimal watershed discharge allocation result.

[0050] In an optional embodiment of the present invention, step S4 uses the KKT condition to transform the two-layer robust optimization model of watershed discharge allocation into a single-layer optimization model to obtain the optimal watershed discharge allocation result, including: The KKT conditional method is used to transform the lower-level objective function and constraint conditions, and the original two-level optimization model is transformed into a single-level optimization model; Clarify the domain of the single-layer optimization model; Randomly select the initial optimal solution within the model domain and record it as ; Use gradient descent to find the optimal solution Iterate and finally get the optimal solution of the model ; Add fuzzy sets to obtain a series of optimal solutions of the optimization model under different uncertainty levels , to achieve the optimal allocation of river basin pollution discharge rights under uncertain environments.

[0051] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0052] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0053] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0054] Specific embodiments are used in the present invention to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core ideas. At the same time, for those skilled in the art, according to the ideas of the present invention, there may be changes in the specific implementation methods and application scopes. In summary, the contents of this specification should not be understood as limiting the present invention.

[0055] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and it should be understood that the scope of protection of the present invention is not limited to such specific descriptions and embodiments. Those skilled in the art can make various other specific variations and combinations based on the technical teachings disclosed in the present invention without departing from the essence of the present invention, and such variations and combinations are still within the scope of protection of the present invention.

Claims

1. A data-driven basin discharge allocation optimization method, characterized in that: The following steps are involved: Obtain basin pollution data; The water environment carrying index is determined based on the basin pollution data. A two-level optimization model for basin pollution allocation is constructed with minimizing pollution control costs as the upper level goal and maximizing pollution allocation benefits as the lower level goal. The uncertainty set of water environment carrying index is introduced into the two-layer optimization model of watershed discharge allocation to construct a two-layer robust optimization model of watershed discharge allocation. The KKT condition is used to transform the two-layer robust optimization model of basin discharge allocation into a single-layer optimization model, and the optimal basin discharge allocation result is obtained.

2. The data-driven basin discharge optimization method according to claim 1 is characterized in that: The water environment carrying index is determined based on the basin pollution data as follows: in, x ik Assigned to the region i pollutants k The amount of sewage discharged, λ ik For the region i Pollutants k The governance ratio, M A collection of regions.

3. The data-driven basin discharge allocation optimization method according to claim 2 is characterized in that: With minimizing pollution control costs as the upper level goal and maximizing pollution discharge allocation benefits as the lower level goal, a two-level optimization model for watershed pollution discharge allocation is constructed, including: Taking minimizing pollution control costs as the upper-level goal, and taking ecological compensation cost constraints, water environment carrying index constraints, discharge demand constraints, emission reduction rate constraints, and parameter non-negativity constraints as constraints, the upper-level model is constructed. The lower-level model is constructed with the maximization of pollution allocation benefits as the lower-level goal, and with the constraints that the pollutant emissions of each polluting enterprise do not exceed the pollutant allocation amount of the region to which it belongs, the pollutant emissions of each polluting enterprise do not exceed the minimum and maximum demand amounts, and the non-negative parameters. A two-layer optimization model for basin discharge allocation is constructed based on the upper model and the lower model.

4. The data-driven basin discharge allocation optimization method according to claim 3 is characterized in that: Taking minimizing pollution control costs as the upper-level goal, and taking ecological compensation cost constraints, water environment carrying index constraints, discharge demand constraints, emission reduction rate constraints, and parameter non-negativity constraints as constraints, the upper-level model is constructed, specifically: ； st in, F For pollution control costs, CF For ecological compensation costs, p jk For polluting enterprises j Acquisition of pollutants k The cost of emissions, y ijk For the region i Allocated to polluting enterprises j pollutants k emissions, c ik For the region i Treating pollutants k the cost, x ik Assigned to the region i pollutants k emissions, K is a collection of pollutants, M is a set of regions, N For polluting enterprises, For the region i Contribution coefficient of participating in ecological compensation, For the region i The responsibility coefficient, is the ecological compensation accounting value, is the upstream region collection, is the downstream area collection, is the environmental carrying index of the watershed for pollutant k, For the region i pollutants k The lowest emissions, For the region i pollutants k Baseline emissions, For the region i pollutants k The minimum emission reduction rate, For the region i pollutants k the highest reduction rate.

5. The data-driven basin discharge allocation optimization method according to claim 4 is characterized in that: Taking maximizing the benefits of pollution discharge allocation as the lower-level goal, and taking the pollutant emissions of each polluting enterprise not exceeding the pollutant allocation amount of the region to which it belongs, the pollutant emissions of each polluting enterprise not exceeding the minimum and maximum requirements, and the non-negative parameters as constraints, the lower-level model is constructed, specifically: st Among them, max is the maximum value function, f i For the region i The pollution discharge configuration income, For the region i Medium-sized polluting enterprises j pollutants k Emission benefits, y ijk For the region i Allocated to polluting enterprises j pollutants k emissions, p jk For polluting enterprises j Acquisition of pollutants k The cost of emissions, is the ecological compensation accounting value, x ik Assigned to the region i pollutants k emissions, K is a collection of pollutants, N For polluting enterprises, For the region i Medium-sized polluting enterprises j Emission of pollutants k The minimum demand, For the region i Medium-sized polluting enterprises j Emission of pollutants k The highest demand.

6. The data-driven basin discharge allocation optimization method according to claim 5 is characterized in that: The uncertainty set of water environment carrying index is introduced into the two-level optimization model of watershed discharge allocation as follows: The water environment carrying index constraint is transformed into uncertainty constraint, and the moment information fuzzy set of the water environment carrying index is used to represent the disturbance of uncertainty parameters. The moment information fuzzy set of water environment carrying index is reconstructed into a solvable event-type fuzzy set.

7. The data-driven basin discharge allocation optimization method according to claim 6 is characterized in that: The disturbance of uncertainty parameters is expressed using the moment information fuzzy set of the water environment carrying index as follows: in, is the moment information fuzzy set, is a random variable that obeys the joint probability distribution P, is a random variable The probability distribution family in K-dimensional real space, is a random variable under distribution P The mean of is the mean of the variance under distribution P, is the covariance mean under distribution P, is a random variable under distribution P Satisfy its upper and lower limits The mean of is the mean vector of random variables, is the mean water environment carrying index of pollutant k, is the covariance matrix of the random variables, is the standard deviation of the random variable, is a unit vector, T represents transpose, is the lower bound vector of the random variable distribution, is the upper bound vector of the random variable distribution, K A collection of pollutants.

8. The data-driven basin discharge allocation optimization method according to claim 7 is characterized in that: The moment information fuzzy set of the water environment carrying index is reconstructed into a solvable event-type fuzzy set as follows: in, is an event-type fuzzy set, For random events, is an auxiliary variable, representing a random variable and its mean The difference, is a random variable under the condition of the occurrence of type 1 random event The mean of Under the condition of the occurrence of type 1 random event Regarding the mean value of k for different pollutants, is a random variable under the condition of the occurrence of type 1 random event The covariance of Under the condition of the occurrence of type 1 random event and Belong to the set The mean probability of is a random variable under distribution P The mean that satisfies its lower and upper bounds.

9. The data-driven basin discharge allocation optimization method according to claim 8 is characterized in that: The specific steps of constructing a two-layer robust optimization model for basin discharge allocation are as follows: 。 10. The data-driven basin discharge allocation optimization method according to claim 9 is characterized in that: The KKT condition is used to transform the two-layer robust optimization model of basin discharge allocation into a single-layer optimization model, and the optimal basin discharge allocation result is obtained, including: The KKT conditional method is used to transform the lower-level objective function and constraint conditions, and the original two-level optimization model is transformed into a single-level optimization model; Clarify the domain of the single-layer optimization model; Randomly select the initial optimal solution within the model domain and record it as ; Use gradient descent to find the optimal solution Iterate and finally get the optimal solution of the model ; Add fuzzy sets to obtain a series of optimal solutions of the optimization model under different uncertainty levels , to achieve the optimal allocation of river basin pollution discharge rights under uncertain environments.

Images