A self-organizing optimization method and system for reducing nitrogen and phosphorus non-point source pollution loads in watersheds

Abstract

This invention provides a self-organizing optimization method and system for reducing nitrogen and phosphorus non-point source pollution in a watershed, comprising: Step 1: dividing the study watershed into sub-units; Step 2: determining the set of feasible nitrogen and phosphorus non-point source pollution reduction measures for the study watershed, and statistically investigating the feasible scale of each measure within each sub-unit; Step 3: statistically investigating the unit scale cost and total nitrogen and total phosphorus pollution load reduction rate of each measure in each sub-unit; Step 4: determining the upper and lower limits of the remaining rates of total nitrogen and total phosphorus pollution load based on water environment protection objectives; Step 5: using the total cost of nitrogen and phosphorus non-point source pollution control in the watershed and the remaining rates of total nitrogen and phosphorus pollution load in the sub-units as constraints, solving the optimal combination of measures based on an adaptive genetic algorithm. This invention solves the problem of how to maximize the reduction of nitrogen and phosphorus pollution load in a watershed by differentially configuring the combination of measures in different spatially heterogeneous sub-units under the dual constraints of cost control and water environment protection objectives.

Description

Technical Field

[0001] This invention belongs to the field of water conservancy engineering technology, specifically relating to a self-organizing optimization method and system for reducing nitrogen and phosphorus non-point source pollution loads in watersheds. Background Technology

[0002] Watershed non-point source pollution is complex, diverse in type, widely spatially distributed, and exhibits significant heterogeneity. Different control measures are required for different non-point source pollution characteristics. Single control technologies often suffer from high implementation costs, unstable pollution control effects, or difficulty in achieving expected goals. Therefore, in practice, a combination of multiple technologies is usually adopted for control.

[0003] Watershed-level nitrogen and phosphorus non-point source pollution control requires a holistic consideration of both control costs and water quality protection objectives. Current research often employs multi-objective optimization methods to seek the optimal combination of control measures at the entire watershed scale. However, these methods have significant limitations: they neglect the spatial heterogeneity of different regions within the watershed (such as sub-basins, administrative regions, and water function zones) in terms of hydrological processes, pollution source structure, land use, topography, and other natural attributes. This leads to significant differences in the implementation costs and pollutant load reduction efficiencies of the same control measures in different regions. This homogenized approach fails to accurately reflect the natural constraints of watershed pollution control, resulting in optimized solutions that may be ineffective or cost-efficient in practical applications, and thus cannot provide spatially differentiated and precise control decision support for complex watershed systems. Summary of the Invention

[0004] This invention aims to overcome the aforementioned shortcomings of existing technologies and solve the technical problem of how to scientifically configure combinations of nitrogen and phosphorus non-point source pollution control measures in different natural geographical units within a watershed, under the dual constraints of limited treatment costs and water environment protection goals, in order to maximize the reduction of nitrogen and phosphorus pollution load across the entire watershed. To this end, this invention provides a self-organizing optimization method and system for watershed nitrogen and phosphorus non-point source pollution load reduction measures based on an adaptive genetic algorithm. This method fully considers the impact of watershed spatial heterogeneity on the cost and effectiveness of treatment measures and strictly follows the natural laws of water environment capacity when setting constraints.

[0005] To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0006] A self-organizing optimization method for reducing nitrogen and phosphorus non-point source pollution loads in a watershed includes the following steps:

[0007] Step 1: Divide the study watershed into sub-units;

[0008] Step 2: Based on the characteristics of nitrogen and phosphorus non-point source pollution in the study watershed, determine the set of feasible nitrogen and phosphorus non-point source pollution load reduction measures for the study watershed, and conduct statistical surveys to obtain the feasible scale of each measure within each sub-unit;

[0009] Step 3: Determine the unit scale cost of each measure in each sub-unit based on the construction content of each measure, and obtain the expected reduction rate of total nitrogen and total phosphorus pollution load;

[0010] Step 4: Determine the upper and lower limits of the residual rates of total nitrogen and total phosphorus pollution loads based on water environment protection objectives;

[0011] Step 5: Using the maximum reduction rate of pollutants in the watershed as the fitness function, and with the total cost of the watershed and the remaining rates of total nitrogen and total phosphorus pollution loads in the sub-units as constraints, use an adaptive genetic algorithm to solve for the optimal combination of measures, and output the optimal combination of measures and its corresponding remaining rates of total nitrogen and total phosphorus pollution loads.

[0012] Furthermore, in step 1, the sub-units are divided according to different research objectives, such as sub-basins, administrative regions, water function zones, and water environment function zones within the research watershed.

[0013] Furthermore, the set of feasible nitrogen and phosphorus non-point source pollution load reduction measures for the watershed studied in step 2 is as follows:

[0014] Based on the characteristics of nitrogen and phosphorus non-point source pollution in the watershed, as well as the topography, hydrology, and water resources, various measures suitable for reducing non-point source pollution within the watershed were selected, forming a set of nitrogen and phosphorus non-point source pollution load reduction measures.

[0015] Furthermore, the statistical survey in step 2 to obtain the feasible scale of each measure within each sub-unit includes:

[0016] Within each sub-unit, the feasibility scale of each measure in the set of feasible nitrogen and phosphorus non-point source pollution load reduction measures is investigated. If a single measure is not required or cannot be implemented in a certain sub-unit, then the feasibility scale of this single measure in this unit is 0.

[0017] Furthermore, in step 3, the unit cost of each measure in each sub-unit is estimated based on the construction content, and the expected pollution load reduction rate is obtained through literature or experimental methods.

[0018] Furthermore, the pollution load surplus rate of each sub-unit in step 4 is given by the following function:

[0019]

[0020] In the formula, i For the measure number, j For sub-unit number, m For the number of measures, n This represents the number of sub-units. , Sub-units jTotal phosphorus and total nitrogen load residual rates, R n ( i , j ) TP , R n ( i , j ) TN The first i After the implementation of the first measure j Total phosphorus and total nitrogen load reduction rates for each subunit.

[0021] The upper and lower limits of the residual rates of total nitrogen and total phosphorus pollution loads are given by the following formulas:

[0022]

[0023] In the formula, P TP 、P TN These represent the current total phosphorus and total nitrogen pollution loads in the watershed, in t / a. Q TP 、Q TN These represent the remaining water environmental capacity for total phosphorus and total nitrogen in the watershed after deducting point source emissions, in t / a. W TP 、W TN The theoretical water environmental capacity for total phosphorus and total nitrogen in the watershed, in t / a; , These represent the upper and lower limits of the total phosphorus pollution load surplus rate, respectively. , These represent the upper and lower limits of the total nitrogen pollution load surplus rate, respectively.

[0024] Furthermore, the constraint conditions are as follows:

[0025]

[0026] In the formula, Total_cost To study the total cost of the watershed, Budget For total cost budget;

[0027] Total_cost Calculate using the following formula:

[0028]

[0029] In the formula, Cost ( i , j ) is the first i The measure in the j Unit-scale cost of implementation in each sub-unit; Scale (i , j ) is the first i The measure in the j The implementation scale of each sub-unit; Sol ( i , j) For the first i The measure in the j A binary variable indicating whether a subunit is implemented or not.

[0030] Further, step 5 includes:

[0031] Step 5.1 The implementation status of the nitrogen and phosphorus non-point source pollution load reduction measures combination scheme set of each sub-unit is encoded in binary, with 1 representing implementation and 0 representing non-implementation, forming a two-dimensional array containing the implementation status of each measure in each sub-unit, which serves as the watershed nitrogen and phosphorus non-point source pollution load reduction measures combination scheme.

[0032] A two-dimensional array of a specified number of watershed nitrogen and phosphorus non-point source pollution load reduction measures combinations is generated using a random function and used as the initial population for the genetic algorithm.

[0033] Step 5.2 Enter the main program of the adaptive genetic algorithm to calculate the fitness value of each individual in the population; the fitness function is calculated using the following formula:

[0034]

[0035] In the formula, Fitness For each individual, the fitness value N residuals (1) TP , N residuals (2) TP , N residuals (n) TP For the 1st, 2nd, n Total phosphorus surplus rate of each sub-unit N residuals (1) TP , N residuals (2) TP , N residuals ( n ) TP For the 1st, 2nd, n Total nitrogen surplus rate of each subunit;

[0036] Genetic algorithms employ roulette wheel selection, single-point crossover, and single-point mutation to form a new generation of population;

[0037] In the selection process, roulette wheel selection is used. The probability that each individual's gene will be selected for the next generation is proportional to its fitness value, calculated as follows:

[0038]

[0039] In the formula, Fitness k The first in the population k The fitness value of each individual;

[0040] In the process of gene crossover, individuals in the population are first paired sequentially, and the probability of gene crossover between each pair of individuals is correlated with the average fitness of these two individuals; the specific calculation of the probability of gene crossover between two individuals is as follows:

[0041]

[0042] In the formula, P 交叉 The probability of gene crossover between two individuals. P 1 represents the maximum probability of gene crossover, a value specified by the user; Fitness max and Fitness min These represent the maximum and minimum fitness values ​​of individuals in the current population. Fitness ave The average fitness of the two individuals where gene crossover occurred;

[0043] The probability of gene mutation for each individual is calculated as follows:

[0044]

[0045] In the formula, P 变异 This represents the probability of gene mutation in the current individual. P 2 represents the maximum probability of gene mutation, a value specified by the user. Fitness This represents the fitness value of the individual.

[0046] The new generation population undergoes selection, crossover, and mutation iterations until a specified number of iterations, after which the iteration terminates, yielding the current optimal combination of measures.

[0047] Step 5.3 Output the current optimal combination of measures in a two-dimensional array, and output the total cost, maximum benefit and corresponding pollution load surplus rate of the scheme.

[0048] Furthermore, step 5 sets four parameters: the number of individuals in the initial population. k Number of population iterations g Maximum crossover probability P 1. Maximum mutation probabilityP 2. Input by the user.

[0049] On the other hand, the present invention provides a self-organizing optimization system for reducing nitrogen and phosphorus non-point source pollution loads in watersheds, comprising:

[0050] Sub-unit partitioning module: This is used to partition the study watershed into sub-units;

[0051] Measures set acquisition module: It is used to determine the set of nitrogen and phosphorus non-point source pollution load reduction measures that can be implemented in the study watershed based on the characteristics of non-point source pollution in the study watershed, and to obtain the implementable scale of each measure in each sub-unit through statistical survey;

[0052] Data acquisition module: It is used to acquire the unit cost and pollution load reduction rate of each measure in each sub-unit;

[0053] The constraint construction module is used to construct total cost constraints based on the unit cost, feasible scale, and total project budget of each measure in each sub-unit, and to construct total nitrogen and total phosphorus reduction rate constraints based on the watershed water quality protection target.

[0054] The solution module uses the maximum pollutant reduction rate of each combination of treatment measures as the fitness value, and studies the total cost of the watershed and the pollution load surplus rate of the sub-unit as constraints. It solves the optimal combination of measures based on an adaptive genetic algorithm and outputs the total cost of the watershed and the corresponding pollution load surplus rate of the optimal combination of measures.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] 1. Refined Spatial Optimization: This invention divides the watershed into sub-units (such as sub-watersheds and functional zones) that reflect natural geographical characteristics or management needs, and considers the cost differences and nitrogen and phosphorus load reduction efficiency differences of the same treatment measures in different sub-units, thus achieving a refined and differentiated spatial configuration of treatment measures. This solves the technical problem that existing watershed-wide homogeneous optimization methods cannot adapt to the spatial heterogeneity within the watershed. It effectively addresses the problems of excessive measures and resource misallocation and waste that may exist in traditional experience-based or extensive planning, enabling limited treatment resources to achieve the maximum pollution control technology effect and significantly improving the accuracy of resource allocation and the potential effectiveness of pollution control.

[0057] 2. Multi-objective Collaborative Optimization and Water Quality Target Assurance: This invention uses the maximum reduction rate of watershed pollutant load as the core optimization objective, while simultaneously treating the total cost of treatment and water environment protection targets as rigid constraints. A genetic algorithm automatically identifies and eliminates redundant or inefficient combinations of treatment measures. This model construction method strictly follows the natural laws of pollutant migration and transformation and water environment capacity, ensuring that the optimized scheme can effectively meet the water quality protection requirements of different regions while maintaining controllable costs, ultimately achieving the dual technical effects of cost control and water quality improvement. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0059] Figure 1 is a flowchart of the method of the present invention. Detailed Implementation

[0060] To make the above-mentioned objectives, features, and advantages of this application more apparent and understandable, the specific embodiments of this application are described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0061] Example 1

[0062] like Figure 1 As shown in the figure, this invention provides a self-organizing optimization method for reducing nitrogen and phosphorus non-point source pollution loads in watersheds, comprising the following steps:

[0063] Step 1: Divide the study watershed into sub-units;

[0064] In this invention, a sub-unit refers to a sub-basin, administrative region, water function zone, water environment function zone, etc.

[0065] Step 2: Based on the characteristics of nitrogen and phosphorus non-point source pollution in the study watershed, determine the set of feasible nitrogen and phosphorus non-point source pollution load reduction measures for the study watershed, and conduct statistical surveys to obtain the feasible scale of each measure within each sub-unit;

[0066] In this invention, step 2, statistically investigating and obtaining the feasible scale of each measure within each sub-unit, means investigating and obtaining the feasible scale of each measure within the set of feasible nitrogen and phosphorus non-point source pollution load reduction measures within each sub-unit. If a certain measure is not required or cannot be implemented in a certain sub-unit, then the feasible scale of this measure in this unit is 0.

[0067] Step 3: Determine the unit scale cost of each measure in each sub-unit based on the construction content of each measure, and obtain the expected reduction rate of total nitrogen and total phosphorus pollution load based on literature or experiments;

[0068] In this invention, the unit cost of each measure in each sub-unit can be estimated through the construction content, and the expected pollution load reduction rate can be obtained through literature or experimental methods.

[0069] Step 4: Determine the upper and lower limits of the residual rates of total nitrogen and total phosphorus pollution loads based on water environment protection objectives;

[0070] In this invention, the pollution load surplus rate of each sub-unit is given by the following function:

[0071] Sub-unit pollution load remaining rate function for:

[0072]

[0073] In the formula, i For the measure number, j For sub-unit number, m For the number of measures, n This represents the number of sub-units. , Subunit j Total phosphorus and total nitrogen load residual rates, R n ( i , j ) TP , R n ( i , j ) TN For the first i After the implementation of the first measure j Total phosphorus and total nitrogen load reduction rates for each subunit.

[0074] The upper and lower limits of the residual nitrogen and total phosphorus pollution load rates are given by the following formulas:

[0075]

[0076] In the formula, X 1 TP , X 2 TPThese are the upper and lower limits of the total phosphorus pollution load surplus rate; X 1 TN , X 2 TN These are the upper and lower limits of the total phosphorus pollution load surplus rate; P TP 、P TN The current total phosphorus and total nitrogen pollution loads (t / a) of the watershed are given. Q TP 、Q TN The remaining water environmental capacity (t / a) for total phosphorus and total nitrogen after deducting point source emissions in the watershed. W TP 、W TN The theoretical water environmental capacity (t / a) for total phosphorus and total nitrogen in the watershed.

[0077] Step 5: Using the maximum reduction rate of pollutants in the watershed as the fitness function, and with the total cost of the watershed and the remaining rates of total nitrogen and total phosphorus pollution loads in the sub-units as constraints, use an adaptive genetic algorithm to solve for the optimal combination of measures, and output the optimal combination of measures and its corresponding remaining rates of total nitrogen and total phosphorus pollution loads.

[0078] In this invention, the constraints are given by the following formula:

[0079]

[0080] In the formula, Total_cost To study the total cost of the watershed, Budget This is the total cost budget.

[0081] Total cost Total_cost Calculate using the following formula:

[0082]

[0083] In the formula, Cost ( i , j ) is the first i The measure in the j Unit-scale cost of implementation in each sub-unit; Scale ( i , j ) is the first i The measure in the j The implementation scale of each sub-unit; Sol ( i , j) For the first i The measure in the j Whether or not the sub-units are implemented.

[0084] In this invention, the implementation status of the nitrogen and phosphorus non-point source pollution load reduction measures combination scheme set of each sub-unit is first encoded using binary code, with 1 representing implementation and 0 representing non-implementation, forming a two-dimensional array containing the implementation status of each measure in each sub-unit, which serves as the watershed nitrogen and phosphorus non-point source pollution load reduction measures combination scheme.

[0085] A two-dimensional array of a specified number of watershed nitrogen and phosphorus non-point source pollution load reduction measures combinations is generated using a random function and used as the initial population for the genetic algorithm.

[0086] Upon entering the main program of the adaptive genetic algorithm, the fitness value of each individual in the population is first calculated; the fitness function is calculated using the following formula:

[0087]

[0088] In the formula, Fitness For each individual, the fitness value N residuals( 1) TP , N residuals (2) TP , N residuals (n) TP For the 1st, 2nd, n Total phosphorus surplus rate of each sub-unit N residuals (1) TP , N residuals (2) TP , N residuals ( n ) TP For the 1st, 2nd, n Total nitrogen surplus rate of each subunit.

[0089] Genetic algorithms employ roulette wheel selection, single-point crossover, and single-point mutation to form a new generation of population.

[0090] In the selection process, roulette wheel selection is used. The probability of each individual's gene being selected to enter the next generation is proportional to its fitness value, ensuring that genes from individuals with higher fitness values ​​are preserved as much as possible during inheritance. The calculation is as follows:

[0091]

[0092] In the formula, Fitness k The first in the population k The fitness value of each individual.

[0093] In the process of gene crossover, individuals in the population are first paired sequentially. The probability of gene crossover between each pair of individuals is related to the average fitness of these two individuals. Individuals with higher average fitness have a lower probability of gene crossover, ensuring that individuals with higher fitness values ​​are preserved as much as possible during inheritance, while individuals with lower fitness can change their fitness through a greater probability of gene crossover. The specific calculation of the probability of gene crossover between two individuals is as follows:

[0094]

[0095] In the formula, P 交叉 The probability of gene crossover between two individuals. P 1 represents the maximum probability of gene crossover, a value specified by the user. Fitness max and Fitness min These represent the maximum and minimum fitness values ​​of individuals in the current population. Fitness ave This represents the average fitness of the two individuals where gene crossover occurred.

[0096] In the process of gene mutation, the probability of gene mutation for each individual is correlated with that individual's fitness. Individuals with higher fitness values ​​have a lower probability of gene mutation, ensuring that genes from individuals with higher fitness values ​​are preserved as much as possible during inheritance. Conversely, genes from individuals with lower fitness values ​​can have their fitness altered through a greater probability of gene mutation. The probability of gene mutation for each individual is calculated as follows:

[0097]

[0098] In the formula, P 变异 This represents the probability of gene mutation in the current individual. P 2 represents the maximum probability of gene mutation, a value specified by the user. Fitness This is the fitness value of the individual.

[0099] The new generation population undergoes selection, crossover, and mutation iterations until a specified number of iterations, after which the iteration terminates, yielding the current optimal combination of measures.

[0100] Output the current optimal combination of measures as a two-dimensional array, along with the total cost of the solution and its corresponding pollution load surplus rate.

[0101] Furthermore, this method sets four parameters: the number of individuals in the initial population. k Number of population iterations g Maximum crossover probability P 1. Maximum mutation probability P2. User input is possible. A larger population size and more iterations result in a stronger global search capability, but also increased computational complexity. Appropriately increasing the maximum crossover probability is recommended. P 1 and maximum mutation probability P 2. It can improve the diversity of individuals in the population and avoid getting trapped in local optima. However, an excessively high maximum crossover probability will slow down the convergence speed of the solution, and an excessively high maximum mutation probability will make the genetic process too random. Users need to set the parameters according to the complexity of the watershed non-point source reduction problem and the limitations of computing resources. k , g , P 1. P 2.

[0102] Specific embodiments of the present invention are as follows:

[0103] Assuming watershed A has 5 sub-units, five nitrogen and phosphorus non-point source pollution load reduction measures were selected based on the location within the watershed: 1) riparian vegetation buffer zone; 2) ecological ditches; 3) soil testing and fertilization; 4) conversion of sloping fields to stone-lined terraces; and 5) small-scale biogas digesters. Non-point source pollution load reduction mainly considers total nitrogen and total phosphorus. According to water environment protection targets, the residual rate of total nitrogen load in the watershed needs to be controlled between 20% and 50%, and the residual rate of total phosphorus load needs to be controlled between 20% and 40%. The cost is controlled at 300 million yuan, and the initial population size for the genetic algorithm is [number of individuals]. k The maximum crossover probability is set to 5000, the number of iterations g is set to 100, and the maximum crossover probability is set to 100. P 1 represents 50%, the maximum mutation probability. P 2 is 50%.

[0104] The implementation cost, feasible scale, total nitrogen reduction rate, and total phosphorus reduction rate of each measure within the five sub-units are shown in Tables 1 to 4.

[0105] Table 1. Implementation cost of each governance measure in each sub-unit

[0106]

[0107] Table 2 Implementation Scale of Each Governance Measures in Each Sub-unit

[0108]

[0109] Table 3 Total nitrogen surface source load reduction rate for each subunit after implementation of each treatment measure.

[0110]

[0111] Table 4. Total phosphorus non-point source load reduction rate for each sub-unit after the implementation of each treatment measure.

[0112]

[0113] Based on the set parameters, the adaptive genetic algorithm main program is run, and the optimized implementation plan results are shown in Table 5, where 1 represents implementing the measure and 0 represents not implementing the measure.

[0114] Table 5 Combinations of Nitrogen and Phosphorus Non-point Source Pollution Load Reduction Measures for Watershed Sub-units

[0115]

[0116] The total cost of this combination of measures is RMB 298.783 million. The total nitrogen non-point source load surplus rates of sub-units 1 to 5 are 30.69%, 32.40%, 39.28%, 35.36%, and 28.60%, respectively; the total phosphorus non-point source load surplus rates of sub-units 1 to 5 are 33.04%, 33.88%, 37.39%, and 34.32%, respectively, which meet the water environment management and protection objectives and cost constraints.

[0117] Example 2

[0118] This embodiment provides a self-organizing optimization system for reducing nitrogen and phosphorus non-point source pollution loads in a watershed, including:

[0119] Sub-unit division module: It is used to divide the study watershed into sub-units;

[0120] Measures set acquisition module: It is used to determine the set of nitrogen and phosphorus non-point source pollution load reduction measures that can be implemented in the study watershed based on the characteristics of non-point source pollution in the study watershed, and to obtain the implementable scale of each measure in each sub-unit through statistical survey;

[0121] Data acquisition module: It is used to acquire the unit scale cost, unit scale benefit, and pollution load reduction rate of each measure in each sub-unit;

[0122] Function building module: It is used to build the total revenue function, the sub-unit pollution load surplus rate function and the penalty function based on the unit size cost, unit size benefit and pollution load reduction rate of each measure in each sub-unit and the implementable scale of each measure in each sub-unit;

[0123] The solution module uses the total benefit value of each combination of governance measures as the fitness value, and with the total cost of the watershed and the pollution load surplus rate of the sub-unit as constraints, it solves for the optimal combination of measures based on the genetic algorithm, and outputs the total cost of the watershed, the maximum total benefit value, and the corresponding pollution load surplus rate of the optimal combination of measures.

[0124] The above description is merely a preferred embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

[0125] It should be understood that any parts not described in detail in this specification belong to the prior art.

[0126] It should be understood that the above description of the preferred embodiments is quite detailed, but it should not be considered as a limitation on the scope of protection of this invention. Those skilled in the art, under the guidance of this invention, can make substitutions or modifications without departing from the scope of protection of the claims of this invention, and all such substitutions or modifications fall within the scope of protection of this invention. The scope of protection of this invention should be determined by the appended claims.

Claims

1. A self-organizing optimization method for reducing nitrogen and phosphorus non-point source pollution loads in a watershed, characterized in that, Includes the following steps: Step 1: Divide the study watershed into sub-units; Step 2: Based on the characteristics of nitrogen and phosphorus non-point source pollution in the study watershed, determine the set of feasible nitrogen and phosphorus non-point source pollution load reduction measures for the study watershed, and conduct statistical surveys to obtain the feasible scale of each measure within each sub-unit; Step 3: Determine the unit scale cost of each measure in each sub-unit based on the construction content of each measure, and obtain the expected total nitrogen pollution load reduction rate and total phosphorus pollution load reduction rate; Step 4: Based on the water environment protection objectives, determine the upper and lower limits of the total nitrogen pollution load surplus rate and the total phosphorus pollution load surplus rate; the pollution load surplus rate of each sub-unit is given by the following function: In the formula, i For the measure number, j For sub-unit number, m For the number of measures, n This represents the number of sub-units. , Sub-units j Total phosphorus load surplus and total nitrogen load surplus R n ( i , j ) TP , R n ( i , j ) TN The first i After the implementation of the first measure j Total phosphorus load reduction rate and total nitrogen load reduction rate of each subunit; The upper and lower limits of the total nitrogen pollution load surplus rate and the total phosphorus pollution load surplus rate are given by the following formulas: In the formula, P TP 、P TN These represent the current total phosphorus pollution load and total nitrogen pollution load of the watershed, respectively. Q TP 、Q TN These refer to the remaining water environmental capacity for total phosphorus and total nitrogen after deducting point source emissions from the watershed. W TP 、W TN The theoretical water environmental capacity for total phosphorus and total nitrogen in the watershed; , These represent the upper and lower limits of the total phosphorus pollution load surplus rate, respectively. , These represent the upper and lower limits of the total nitrogen pollution load surplus rate, respectively. Step 5: Using the maximum pollutant reduction rate of the watershed as the fitness function, and with the total watershed cost and the remaining rates of total nitrogen and total phosphorus pollution loads in sub-units as constraints, use an adaptive genetic algorithm to solve for the optimal combination of measures, and output the optimal combination of measures and its corresponding remaining rates of total nitrogen and total phosphorus pollution loads; the constraints are: In the formula, Total cost To study the total cost of the watershed, Budget For total cost budget; Total cost Calculate using the following formula: In the formula, Cost ( i , j ) is the first i The measure in the j Unit-scale cost of implementation in each sub-unit; Scale ( i , j ) is the first i The measure in the j The implementation scale of each sub-unit; Sol ( i , j) For the first i The measure in the j A binary variable indicating whether a subunit is implemented or not.

2. The self-organizing optimization method for reducing nitrogen and phosphorus non-point source pollution loads in a watershed according to claim 1, characterized in that, In step 1, the sub-units are divided according to different research objectives, such as sub-basins, administrative regions, water function zones, and water environment function zones within the research watershed.

3. The self-organizing optimization method for reducing nitrogen and phosphorus non-point source pollution loads in a watershed according to claim 1, characterized in that, The set of feasible nitrogen and phosphorus non-point source pollution load reduction measures for the watershed studied in step 2 is as follows: Based on the characteristics of nitrogen and phosphorus non-point source pollution in the watershed, as well as the topography, hydrology, and water resources, various measures suitable for reducing non-point source pollution within the watershed were selected, forming a set of nitrogen and phosphorus non-point source pollution load reduction measures.

4. The self-organizing optimization method for reducing nitrogen and phosphorus non-point source pollution loads in a watershed according to claim 1, characterized in that, The statistical survey in step 2 to obtain the feasible scale of each measure within each sub-unit includes: Within each sub-unit, the feasibility scale of each measure in the set of feasible nitrogen and phosphorus non-point source pollution load reduction measures is investigated. If a single measure is not required or cannot be implemented in a certain sub-unit, then the feasibility scale of this single measure in this unit is 0.

5. The self-organizing optimization method for reducing nitrogen and phosphorus non-point source pollution loads in a watershed according to claim 1, characterized in that, In step 3, the unit cost of each measure in each sub-unit is estimated based on the construction content, and the expected pollution load reduction rate is obtained through literature or experimental methods.

6. The self-organizing optimization method for reducing nitrogen and phosphorus non-point source pollution loads in a watershed according to claim 1, characterized in that, Step 5 includes: Step 5.1 The implementation status of the nitrogen and phosphorus non-point source pollution load reduction measures combination scheme set of each sub-unit is encoded in binary, with 1 representing implementation and 0 representing non-implementation, forming a two-dimensional array containing the implementation status of each measure in each sub-unit, which serves as the watershed nitrogen and phosphorus non-point source pollution load reduction measures combination scheme. A two-dimensional array of a specified number of watershed nitrogen and phosphorus non-point source pollution load reduction measures combinations is generated using a random function and used as the initial population for the genetic algorithm. Step 5.2 Enter the main program of the adaptive genetic algorithm to calculate the fitness value of each individual in the population; the fitness function is calculated using the following formula: In the formula, Fitness For each individual, the fitness value N residuals (1) TP , N residuals (2) TP , N residuals (n) TP For the 1st, 2nd, n Total phosphorus surplus rate of each sub-unit N residuals (1) TP , N residuals (2) TP , N residuals ( n ) TP For the 1st, 2nd, n Total nitrogen surplus rate of each subunit; Genetic algorithms employ roulette wheel selection, single-point crossover, and single-point mutation to form a new generation of population; In the selection process, roulette wheel selection is used. The probability that each individual's gene will be selected for the next generation is proportional to its fitness value, calculated as follows: In the formula, Fitness k The first in the population k The fitness value of each individual; In the process of gene crossover, individuals in the population are first paired sequentially, and the probability of gene crossover between each pair of individuals is correlated with the average fitness of these two individuals; the specific calculation of the probability of gene crossover between two individuals is as follows: In the formula, P 交叉 The probability of gene crossover between two individuals. P 1 represents the maximum probability of gene crossover, a value specified by the user; Fitness max and Fitness min These represent the maximum and minimum fitness values ​​of individuals in the current population. Fitness ave The average fitness of the two individuals where gene crossover occurred; The probability of gene mutation for each individual is calculated as follows: In the formula, P 变异 This represents the probability of gene mutation in the current individual. P 2 represents the maximum probability of gene mutation, a value specified by the user. Fitness This represents the fitness value of the individual. The new generation population undergoes selection, crossover, and mutation iterations until a specified number of iterations, after which the iteration terminates, yielding the current optimal combination of measures. Step 5.3 Output the current optimal combination of measures in a two-dimensional array, and output the total cost, maximum benefit and corresponding pollution load surplus rate of the scheme.

7. The self-organizing optimization method for reducing nitrogen and phosphorus non-point source pollution loads in a watershed according to claim 6, characterized in that, Step 5 sets four parameters: the number of individuals in the initial population. k Population iteration count g The highest probability of gene crossover P 1. The highest probability of gene mutation occurring P 2. Input by the user.

8. A self-organizing optimization system for reducing nitrogen and phosphorus non-point source pollution loads in a watershed, characterized in that, include: Sub-unit division module: It is used to divide the study watershed into sub-units; Measures set acquisition module: It is used to determine the set of nitrogen and phosphorus non-point source pollution reduction measures that can be implemented in the study watershed based on the characteristics of nitrogen and phosphorus non-point source pollution in the study watershed, and to obtain the implementable scale of each measure in each sub-unit through statistical survey; Data acquisition module: It is used to acquire the unit cost and nitrogen and phosphorus pollution load reduction rate of each measure in each sub-unit; The constraint construction module is used to construct total cost constraints based on the unit cost, feasible scale, and total project budget of each measure in each sub-unit, and to construct total nitrogen and total phosphorus reduction rate constraints based on the watershed water quality protection target. The solution module uses the maximum pollutant reduction rate of each combination of treatment measures as the fitness value, and studies the total cost of the watershed and the pollution load surplus rate of the sub-unit as constraints. It solves the optimal combination of measures based on an adaptive genetic algorithm and outputs the total cost of the watershed and the corresponding pollution load surplus rate of the optimal combination of measures. The self-organizing optimization system for reducing nitrogen and phosphorus non-point source pollution loads in the watershed is used to execute the steps in the self-organizing optimization method for reducing nitrogen and phosphorus non-point source pollution loads in the watershed as described in any one of claims 1-7.

Images