Regional agricultural planting structure optimization method and device based on constrained multi-objective evolutionary algorithm, equipment and medium

Abstract

The invention relates to a regional agricultural planting structure optimization method and device, equipment and a medium. The method comprises the following steps: respectively calculating and determining a first fitness value of each population individual in a first mixed population and a second fitness value of each population individual in a second mixed population by adopting a first agricultural planting structure optimization algorithm according to a target function and constraint conditions; when a preset iteration condition is met, outputting an individual solution set of the first population and an individual solution set of the second population; and using the individual solution set of the second population as an individual solution set of a third population, iterating the individual solution set of the first population, the individual solution set of the second population and the individual solution set of the third population by using a second agricultural planting structure optimization algorithm, and when the number of iterations reaches a preset number of iterations, obtaining a second agricultural planting structure. And outputting a non-dominated solution in the first population as an optimal agricultural planting structure scheme set. The agricultural planting structure can be balanced between water saving and economic benefit improvement.

Description

Technical Field

[0001] The present application relates to the field of agricultural production, and in particular to a method for optimizing regional agricultural planting structure based on a constrained multi-objective evolutionary algorithm, a corresponding device, an electronic device and a computer-readable storage medium. Background Art

[0002] With the rapid development of agricultural mechanization and modernization, while the total grain output continues to grow, it also brings significant pressure to water resource protection and agricultural carbon emissions.

[0003] When optimizing agricultural planting structure, multiple objectives and constraints need to be considered. Multiple objectives are often conflicting with each other. Pursuing one of the objectives often leads to the deterioration of other objectives. The existence of constraints will make the constrained multi-objective evolutionary algorithm easy to fall into the dilemma of local optimality or poor convergence during the optimization process, affecting the quality of the final optimization solution set.

[0004] To sum up, in the existing technology, multiple objectives and constraints need to be considered when optimizing agricultural planting structure. Multiple objectives are often conflicting with each other. Pursuing one of the objectives will often lead to the deterioration of other objectives. The existence of constraints will make the constrained multi-objective evolutionary algorithm easy to fall into the dilemma of local optimality or poor convergence during the optimization process, affecting the quality of the final optimization solution set. In order to solve this problem, the applicant has made corresponding explorations. Summary of the invention

[0005] The purpose of this application is to solve the above-mentioned problems and to provide a regional agricultural planting structure optimization method based on a constrained multi-objective evolutionary algorithm, a corresponding device, an electronic device and a computer-readable storage medium.

[0006] In order to meet the various objectives of this application, this application adopts the following technical solutions:

[0007] A method for optimizing regional agricultural planting structure based on a constrained multi-objective evolutionary algorithm is proposed to meet one of the purposes of this application, including:

[0008] In response to an instruction to optimize the agricultural planting structure of a target area, a decision variable, an objective function, and a constraint condition are obtained, wherein the decision variable is the planting area corresponding to various crops, the objective function includes the total water demand of crops corresponding to various crops, the economic benefit, and the difference between carbon absorption and carbon emission, and the constraint condition includes a constraint on the total cultivated land area, a constraint on the total regional water supply, a constraint on food security, and a non-negative constraint on the difference between carbon absorption and carbon emission;

[0009] Taking the minimum total water requirement of the crops, the maximum economic benefit and the minimum difference between the carbon absorption and the carbon emission as the optimization direction, a preset first agricultural planting structure optimization algorithm is used to randomly generate a first population and a second population, and the first population and the second population are iterated to determine a first offspring population corresponding to the first population and a second offspring population corresponding to the second population, wherein the first agricultural planting structure optimization algorithm is a constrained multi-objective evolutionary algorithm based on a dual-population co-evolution strategy;

[0010] Determine a first mixed population and a second mixed population according to the first population, the second population, the first offspring population, and the second offspring population, respectively, and calculate and determine a first fitness value corresponding to each population individual in the first mixed population and a second fitness value corresponding to each population individual in the second mixed population according to the objective function and the constraint conditions, wherein the population includes a plurality of population individuals, and each population individual represents a planting area corresponding to various crops;

[0011] The above steps are executed repeatedly until the difference between the sum of the second fitness values ​​in the second mixed population and the sum of the fitness values ​​of the second population of the previous generation is less than a preset threshold, and the individual solutions in the second population are all non-dominated solutions, and the individual solution set of the first population and the individual solution set of the second population are output;

[0012] The individual solution set of the second population is used as the individual solution set of the third population, and the preset second agricultural planting structure optimization algorithm is used to iterate the individual solution set of the first population, the individual solution set of the second population and the individual solution set of the third population. When the number of iterations reaches the preset number of iterations, the non-dominated solutions in the first population are output as the optimal agricultural planting structure solution set, wherein the second agricultural planting structure optimization algorithm is a constrained multi-objective evolutionary algorithm based on a dual traction mechanism and a dynamic search attention strategy, and the dual traction mechanism includes a binary tournament selection strategy and a random selection strategy.

[0013] Optionally, the expression of the objective function corresponding to the total amount of water required by crops is:

[0014]

[0015] Among them, f 1 (x) represents the total water requirement of crops, w i represents the water requirement per unit area of ​​the i-th crop, x i represents the unit planting area of ​​the i-th crop, and I represents the total number of crop types;

[0016] The expression of the objective function corresponding to the economic benefit is:

[0017]

[0018] Among them, f 2 (x) represents economic benefit, p i represents the unit output price of the i-th crop, y i represents the yield per unit area of ​​the i-th crop, c i represents the fixed cost per unit area of ​​the i-th crop, which includes any number of labor cost, seed cost, diesel cost, fertilizer cost, pesticide cost and agricultural film cost, x i represents the unit planting area of ​​the i-th crop, and I represents the total number of crop types;

[0019] The expression of the objective function corresponding to the difference between the carbon absorption and the carbon emission is:

[0020]

[0021] Among them, f 3 (x) represents the difference between carbon absorption and carbon emission, a i represents the carbon absorption coefficient per unit yield of the i-th crop, y i represents the yield per unit area of ​​the i-th crop, e i represents the carbon emission coefficient per unit area of ​​the i-th crop, g j represents the emission coefficient per unit usage of the jth production factor, b ij represents the usage per unit area of ​​the jth production factor on the ith crop, where the production factor includes any number of diesel, fertilizer, pesticide and agricultural film, x i represents the unit planting area of ​​the i-th crop, I represents the total number of crop types, and J represents the total number of production factors;

[0022] The expression of the total cultivated land area constraint is:

[0023]

[0024] Among them, S represents the total area of ​​agricultural land in the target area, x i represents the unit planting area of ​​the i-th crop, and I represents the total number of crop types;

[0025] The expression of the regional water supply total amount constraint is:

[0026]

[0027] Where W represents the total water supply for agricultural cultivation in the target area, x irepresents the unit planting area of ​​the i-th crop, and I represents the total number of crop types;

[0028] The expression of the food security constraint is:

[0029]

[0030] Where N represents the annual per capita food demand in the target area, K represents the population of the target area, and y i represents the yield per unit area of ​​the i-th crop, x i represents the unit planting area of ​​the i-th crop, and I represents the total number of crop types;

[0031] The expression for the non-negative constraint of the difference between carbon absorption and carbon emission is:

[0032]

[0033] Among them, a i represents the carbon absorption coefficient per unit area of ​​the i-th crop, y i represents the yield per unit area of ​​the i-th crop, e i represents the carbon emission coefficient per unit area of ​​the i-th crop, g j represents the emission coefficient per unit usage of the i-th production factor, b ij represents the usage per unit area of ​​the jth production factor in the ith crop, x i represents the unit planting area of ​​the i-th crop, I represents the total number of crop types, and J represents the total number of production factors.

[0034] Optionally, the steps of randomly generating a first population and a second population using a preset first agricultural planting structure optimization algorithm, and iterating the first population and the second population to determine a first offspring population corresponding to the first population and a second offspring population corresponding to the second population include:

[0035] Taking the minimum total water requirement of the crops, the maximum economic benefit and the minimum difference between the carbon absorption and carbon emission as the optimization direction, the first agricultural planting structure optimization algorithm is used to initialize the first population, the second population, the population size, the maximum number of iterations, the current number of iterations variable and the preset threshold, wherein the current number of iterations variable is t 1 , the population size is N, and the maximum number of iterations is T 1max， The preset threshold is α, which is used to determine whether the population falls into a stopped state;

[0036] Using a random selection strategy to generate a first mating pool according to the first population, and using simulated binary crossover and polynomial mutation to randomly select parent individuals from the first mating pool to generate offspring individuals, so as to determine the first offspring population;

[0037] Using a random selection strategy to generate a second mating pool according to the second population, and using simulated binary crossover and polynomial mutation to randomly select parent individuals from the second mating pool to generate offspring individuals, so as to determine the second offspring population;

[0038] Merging the first population, the first offspring population and the second offspring population to determine the first mixed population, combining constraint dominance rule sorting with crowding distance sorting to calculate and determine the fitness value corresponding to the first mixed population, and selecting the best multiple population individuals as the first population of the next generation;

[0039] Merging the second population, the first offspring population and the second offspring population to determine the second mixed population, combining non-dominated sorting with crowding distance sorting to calculate and determine the fitness value corresponding to the second mixed population, and selecting the best multiple population individuals as the next generation second population;

[0040] The above steps are executed repeatedly until the difference between the sum of the second fitness values ​​in the second mixed population and the sum of the fitness values ​​of the previous generation of the second population is less than a preset threshold α, and the individual solutions in the second population are all non-dominated solutions, then the individual solution set of the first population and the individual solution set of the second population are output;

[0041] Or until the current iteration number t 1 Reach the maximum number of iterations T 1max , then stop the iteration and output the individual solution set of the first population and the individual solution set of the second population.

[0042] Optionally, the step of using the individual solution set of the second population as the individual solution set of the third population, using a preset second agricultural planting structure optimization algorithm to iterate the individual solution set of the first population, the individual solution set of the second population, and the individual solution set of the third population, and when the number of iterations reaches a preset number of iterations, outputting the non-dominated solutions in the first population as the optimal agricultural planting structure solution set includes:

[0043] Taking the minimization of the total amount of water required by the crops, the maximization of the economic benefits and the minimization of the difference between the carbon absorption and the carbon emission as the optimization direction, the individual solution set of the second population is used as the individual solution set of the third population;

[0044] The preset second agricultural planting structure optimization algorithm is used to initialize the ideal point, the maximum constraint violation value of the second population, the population size, the maximum number of iterations, the current number of iterations, and the minimum range of mating allowed between individuals, and a set of uniform reference vector sets are generated in the target space, where the ideal point is Z * , which represents the minimum value of each objective value in the second population, and the maximum constraint violation value of the second population is ε 0 , the population size is N, and the maximum number of iterations is T 2max , the current iteration number variable is t 2 , the minimum range of mating allowed between individuals is β, and the reference vector set is Λ={λ 1 ,λ 2 ,λ 3 ,...,λ N};

[0045] Calling a preset double traction mechanism, using a binary tournament selection strategy to generate a third mating pool according to the first population, and using simulated binary crossover and polynomial mutation to randomly select parent individuals from the third mating pool to generate offspring individuals, so as to determine a third offspring population;

[0046] Using a random selection strategy to generate a fourth mating pool according to the second population, and using simulated binary crossover and polynomial mutation to randomly select parent individuals from the fourth mating pool to generate offspring individuals, so as to determine a fourth offspring population;

[0047] A dynamic search attention strategy is used to select individuals entering the fifth mating pool and parent individuals participating in mating according to the third population, and simulated binary crossover and polynomial mutation are used to generate offspring individuals according to the parent individuals to determine the fifth offspring population;

[0048] Merge the first population, the third offspring population, the fourth offspring population and the fifth offspring population to determine a third mixed population, combine the constraint dominance rule sorting and the crowding distance sorting to calculate and determine the fitness value corresponding to the third mixed population, and select the best multiple population individuals as the first population of the next generation;

[0049] The second population, the third offspring population, the fourth offspring population and the fifth offspring population are combined to determine the fourth mixed population, and the fitness value corresponding to the fourth mixed population is determined by combining the ε-constraint method with the crowding distance sorting calculation, and the optimal multiple population individuals are selected as the second population of the next generation, wherein the calculation formula of the ε value used in this iteration in the ε-constraint method is expressed as:

[0050]

[0051] Among them, ε(t 2) represents the ε value of this iteration, t 2 Indicates the current iteration number, T 2max represents the maximum number of iterations, F 2 represents the feasible rate of the second population, ε 0 The maximum constraint violation value of the second population at the beginning of the second round of optimization;

[0052] Merging the third population, the third offspring population, the fourth offspring population and the fifth offspring population to obtain a fifth mixed population, and selecting a plurality of population individuals with low constraint violation values ​​and good diversity from the fifth mixed population as the third population of the next generation according to an environmental selection strategy based on a reference vector and a constraint violation value;

[0053] If the current iteration number t 2 Equal to the maximum number of iterations T 2max , stop the iteration, and output the non-dominated solutions in the first population as the optimal agricultural planting structure solution set.

[0054] Optionally, the steps of using a dynamic search attention strategy to select individuals entering the fifth mating pool and parent individuals participating in mating according to the third population, and using simulated binary crossover and polynomial mutation to generate offspring individuals according to the parent individuals to determine the fifth offspring population include:

[0055] Select N in the third population using a binary tournament selection strategy fit individuals enter the fifth mating pool, where N fit The value calculation formula is expressed as:

[0056]

[0057] Among them, N fit represents the number of individuals selected in the third population using the binary tournament selection strategy, t 2 Indicates the current iteration number, T 2max Indicates the maximum number of iterations;

[0058] Use random selection strategy to select NN in the third population fit Individuals enter the fifth mating pool, and the range of mating allowed between two individuals at the current iteration number is calculated and determined, where the range of mating allowed between two individuals at the current iteration number is the Dis value;

[0059] When the Dis value is less than the minimum range β allowed for mating between two individuals, let the Dis value be equal to β. The calculation formula of the Dis value is expressed as:

[0060]

[0061] Among them, the Dis value represents the range of mating allowed between two individuals in the current iteration number, t 2 Indicates the current iteration number, T 2max Indicates the maximum number of iterations;

[0062] Each individual in the fifth mating pool randomly selects a parent individual to mate with from the Dis individuals in the fifth mating pool with the closest Euclidean distance to it, and generates offspring individuals by using simulated binary crossover and polynomial mutation to form a fifth offspring population.

[0063] Optionally, the step of merging the third population, the third offspring population, the fourth offspring population and the fifth offspring population to obtain a fifth mixed population, and selecting a plurality of population individuals with low constraint violation values ​​and good diversity from the fifth mixed population as the third population of the next generation according to an environmental selection strategy based on a reference vector and a constraint violation value comprises:

[0064] Update the ideal point Z according to the target information of the first population and the second population * , assigning the individuals in the first population and the fifth mixed population to N sub-regions through the reference vector set;

[0065] The individual associated with the ith subregion in the first population is denoted as M i , the individual associated with the ith sub-region in the fifth mixed population is denoted as P i ;

[0066] If M i is an empty set and P i is an empty set, select the individual closest to the ith subregion from the first population to enter the third population of the next generation; otherwise, if M i is an empty set and P i is not an empty set, from P i Select the individual with the smallest constraint violation value, and then select g from these individuals te The individual with the smallest value enters the third population of the next generation; otherwise, i Select the individual with the smallest constraint violation value, and then select g from these individuals te The individual with the smallest value enters the third population of the next generation, where g te The value is the function value calculated by the modified Chebyshev method, and its calculation formula is expressed as:

[0067]

[0068] Among them, g te (x|λ,Z * ) represents the g of an individual te Value, λ h represents the hth target of the reference vector, fh (x) represents the h-th fitness value of an individual, Represents the hth fitness value of the ideal point.

[0069] Optionally, the agricultural planting structure scheme set represents an individual solution set constructed by the planting areas corresponding to various crops; the crops include rice crops, wheat crops and cotton crops.

[0070] A regional agricultural planting structure optimization device based on a constrained multi-objective evolutionary algorithm is provided to meet another purpose of the present application, comprising:

[0071] a parameter acquisition module, configured to acquire decision variables, objective functions, and constraints in response to an instruction to optimize the agricultural planting structure of a target area, wherein the decision variables are the planting areas corresponding to various crops, the objective functions include the total water demand of crops corresponding to various crops, economic benefits, and the difference between carbon absorption and carbon emissions, and the constraints include the total cultivated land area constraint, the total regional water supply constraint, the food security constraint, and the difference between carbon absorption and carbon emissions is a non-negative constraint;

[0072] The first planting structure optimization module is configured to minimize the total water requirement of the crops, maximize the economic benefits, and minimize the difference between the carbon absorption and carbon emissions as the optimization direction, randomly generate a first population and a second population using a preset first agricultural planting structure optimization algorithm, iterate the first population and the second population to determine a first offspring population corresponding to the first population and a second offspring population corresponding to the second population, wherein the first agricultural planting structure optimization algorithm is a constrained multi-objective evolutionary algorithm based on a dual-population co-evolutionary strategy;

[0073] A fitness value determination module is configured to respectively determine a first mixed population and a second mixed population according to the first population, the second population, the first offspring population, and the second offspring population, and respectively calculate and determine a first fitness value corresponding to each population individual in the first mixed population and a second fitness value corresponding to each population individual in the second mixed population according to the objective function and the constraint conditions, wherein the population includes a plurality of population individuals, and each population individual represents a planting area corresponding to various crops;

[0074] The individual solution set determination module is configured to execute the above steps cyclically until the difference between the sum of the second fitness values ​​in the second mixed population and the sum of the fitness values ​​of the previous generation second population is less than a preset threshold, and the individual solutions in the second population are all non-dominated solutions, and then output the individual solution set of the first population and the individual solution set of the second population;

[0075] The second planting structure optimization module is configured to use the individual solution set of the second population as the individual solution set of the third population, and adopt a preset second agricultural planting structure optimization algorithm to iterate the individual solution set of the first population, the individual solution set of the second population, and the individual solution set of the third population. When the number of iterations reaches a preset number of iterations, the non-dominated solutions in the first population are output as the optimal agricultural planting structure solution set, wherein the second agricultural planting structure optimization algorithm is a constrained multi-objective evolutionary algorithm based on a dual traction mechanism and a dynamic search attention strategy, and the dual traction mechanism includes a binary tournament selection strategy and a random selection strategy.

[0076] An electronic device provided to meet another purpose of the present application includes a central processing unit and a memory, wherein the central processing unit is used to call and run a computer program stored in the memory to execute the steps of the regional agricultural planting structure optimization method based on a constrained multi-objective evolutionary algorithm described in the present application.

[0077] A computer-readable storage medium is provided to meet another purpose of the present application, which stores a computer program implemented according to the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm in the form of computer-readable instructions. When the computer program is called and executed by a computer, the steps included in the corresponding method are executed.

[0078] Compared with the prior art, the present application aims at the problem that in the prior art, multiple objectives and constraints need to be considered when optimizing agricultural planting structure. Multiple objectives are often conflicting with each other. Pursuing one of the objectives often leads to the deterioration of other objectives. The existence of constraints will make the constrained multi-objective evolutionary algorithm easy to fall into the dilemma of local optimum or poor convergence during the optimization process, affecting the quality of the final optimization solution set. The present application includes but is not limited to the following beneficial effects:

[0079] First, the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm of the present application adopts a dual-population co-evolution strategy in the first round of optimization. This strategy introduces two groups of populations, which evolve separately and influence each other, and can avoid the problem of poor convergence caused by difficulty in crossing the infeasible area during the solution process. Dual-population co-evolution can enhance the diversity of the population and avoid the optimization process from converging to an undesirable solution too early. Therefore, the first round of optimization can effectively cross a large range of infeasible areas, provide a wider search space, and ensure that a more reasonable planting plan is found.

[0080] Secondly, the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm of this application adopts a dual traction mechanism and a dynamic search attention strategy in the second round of optimization. The dual traction mechanism includes a binary tournament selection strategy and a random selection strategy, which helps the algorithm avoid the dilemma of premature convergence and local optimal solution. Specifically, the binary tournament selection strategy ensures the competitiveness between individual solutions and ensures that only high-quality solutions can be selected, while the random selection strategy avoids the population from falling into the limitation of a single choice.

[0081] Thirdly, the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm of this application has enhanced the adaptive ability of the population in the search space by introducing a dynamic search attention strategy. In the early stage of optimization, the algorithm can conduct extensive exploration to find possible optimal solutions; and as the number of iterations increases, the attention gradually turns to local development and focuses more on the nearby areas of potential optimal solutions, thereby accelerating convergence and avoiding falling into local optimality.

[0082] Fourthly, the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm of the present application, through two rounds of optimization, the obtained agricultural planting structure solution set not only has better convergence, that is, it can quickly approach the optimal solution, but also has higher diversity and can provide multiple effective solutions. This diversity means that different agricultural planting plans can be flexibly selected according to different regional needs, climatic conditions and market conditions, so that they have stronger adaptability in practical applications.

[0083] Fifth, the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm of this application, by comprehensively considering multiple factors such as water resource conservation and economic benefits, the obtained agricultural planting structure optimization plan has better sustainable development potential. Through this optimization method, not only can the agricultural production efficiency and income be improved, but also it can ensure that agricultural activities can continue under the background of limited resources and increasing environmental pressure.

[0084] Furthermore, the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm of the present application provides a new solution for the optimization of regional agricultural planting structure, which has significant beneficial effects. Through reasonable algorithm design and multi-objective optimization, the agricultural planting structure can achieve a balance in ensuring water conservation and improving economic benefits. Combined with the dual-population co-evolution strategy and dual traction mechanism, the algorithm can improve convergence and avoid local optimal solutions, thereby improving the quality and practicality of the optimization solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0086] Figure 1 A schematic diagram of the flow of a method for optimizing regional agricultural planting structure based on a constrained multi-objective evolutionary algorithm in an embodiment of the present application;

[0087] Figure 2 A schematic diagram of a constrained multi-objective evolutionary algorithm based on a dual-population co-evolutionary strategy in an embodiment of the present application;

[0088] Figure 3 A schematic diagram of a multi-objective evolutionary algorithm constrained by a dual traction mechanism and a dynamic search attention strategy in an embodiment of the present application;

[0089] Figure 4 This is a schematic diagram of the effect of the mating range between individuals on the search range of offspring in the embodiments of the present application;

[0090] Figure 5 This is a principle block diagram of a device for optimizing regional agricultural planting structure based on a constrained multi-objective evolutionary algorithm in an embodiment of the present application;

[0091] Figure 6 It is a schematic diagram of the structure of the computer device in the embodiment of the present application. DETAILED DESCRIPTION

[0092] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be interpreted as limiting the present application.

[0093] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present application refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0094] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art to which this application belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with those in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined as here.

[0095] It will be understood by those skilled in the art that the "client", "terminal" and "terminal device" used herein include both devices with wireless signal receivers, which are devices with only wireless signal receivers without transmission capabilities, and devices with receiving and transmitting hardware, which are devices with receiving and transmitting hardware capable of two-way communication on a two-way communication link. Such devices may include: cellular or other communication devices such as personal computers, tablet computers, which have single-line displays or multi-line displays or cellular or other communication devices without multi-line displays; PCS (Personal Communications Service, personal communication system), which can combine voice, data processing, fax and / or data communication capabilities; PDA (Personal Digital Assistant, personal digital assistant), which may include a radio frequency receiver, pager, Internet / intranet access, web browser, notepad, calendar and / or GPS (Global Positioning System, global positioning system) receiver; conventional laptop and / or palmtop computers or other devices, which have and / or include a conventional laptop and / or palmtop computer or other device with and / or including a radio frequency receiver. The "client", "terminal" and "terminal device" used herein may be portable, transportable, installed in a vehicle (air, sea and / or land), or suitable for and / or configured to run locally, and / or in a distributed form, at any other location on the earth and / or in space. The "client", "terminal" and "terminal device" used herein may also be a communication terminal, an Internet terminal, a music / video playing terminal, for example, a PDA, a MID (Mobile Internet Device) and / or a mobile phone with a music / video playing function, or a smart TV, a set-top box and other devices.

[0096] The hardware referred to by the names such as "server", "client", and "service node" in this application is essentially an electronic device with the equivalent capabilities of a personal computer. It is a hardware device with the necessary components revealed by the von Neumann principle, such as a central processing unit (including an arithmetic unit and a controller), a memory, an input device, and an output device. The computer program is stored in its memory, and the central processing unit calls the program stored in the external memory into the internal memory for execution, executes the instructions in the program, and interacts with the input and output devices to complete specific functions.

[0097] It should be pointed out that the concept of "server" referred to in this application can also be extended to the case of server clusters. According to the network deployment principle understood by those skilled in the art, the servers should be logically divided. In physical space, these servers can be independent of each other but can be called through interfaces, or integrated into a physical computer or a set of computer clusters. Those skilled in the art should understand this flexibility, and should not use it to restrict the implementation of the network deployment method of this application.

[0098] Unless expressly specified, one or more technical features of the present application can be deployed on a server for implementation and accessed by a client through a remote call to obtain an online service interface provided by the server, or can be directly deployed and run on a client for access.

[0099] The neural network models referenced or may be referenced in this application, unless expressly specified, can be deployed on a remote server and remotely called on the client, or can be deployed and directly called on a client with sufficient device capabilities. In some embodiments, when it runs on the client, its corresponding intelligence can be obtained through transfer learning to reduce the requirements for the client's hardware operating resources and avoid excessive occupation of the client's hardware operating resources.

[0100] Unless explicitly specified, the various data involved in this application can be stored remotely on a server or on a local terminal device, as long as it is suitable for being called by the technical solution of this application.

[0101] Those skilled in the art should be aware that, although the various methods of the present application are described based on the same concept and thus present commonality to each other, unless otherwise specified, these methods can be independently executed. Similarly, for each embodiment disclosed in the present application, they are all proposed based on the same inventive concept, therefore, concepts with the same expression, and concepts that are appropriately changed for convenience despite different expressions, should be understood as equivalent.

[0102] Unless the mutually exclusive relationship between the embodiments to be disclosed in this application is explicitly stated, the relevant technical features involved in each embodiment can be cross-combined to flexibly construct a new embodiment, as long as such combination does not deviate from the creative spirit of this application and can meet the needs of the prior art or solve certain deficiencies in the prior art. Those skilled in the art should be aware of this flexibility.

[0103] See also Figure 1 In one embodiment, the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm of the present application includes:

[0104] Step S10, in response to the instruction to optimize the agricultural planting structure of the target area, obtaining decision variables, objective functions and constraints, wherein the decision variables are the planting areas corresponding to various crops, the objective functions include the total water demand of crops corresponding to various crops, economic benefits and the difference between carbon absorption and carbon emissions, and the constraints include the total cultivated land area constraint, the regional water supply total amount constraint, the food security constraint and the difference between carbon absorption and carbon emissions being a non-negative constraint;

[0105] The regional agricultural planting structure optimization system in the terminal device can obtain decision variables, objective functions and constraints in response to the instruction to optimize the agricultural planting structure of the target area, wherein the decision variables are the planting areas corresponding to various crops, the objective functions include the total water demand of crops corresponding to various crops, economic benefits and the difference between carbon absorption and carbon emissions, and the constraints include the total area of ​​cultivated land constraints, the total regional water supply constraints, food security constraints and the difference between carbon absorption and carbon emissions is a non-negative constraint; wherein the crops include rice crops, wheat crops and cotton crops, and the agricultural planting structure is the configuration and distribution of the types, areas, yields and layouts of crops planted in the target area within a certain period of time. Specifically, it reflects the proportion and planting area of ​​different types of crops in agricultural production. The rationality of the agricultural planting structure has an important impact on agricultural production efficiency, resource utilization, environmental protection and farmers' income.

[0106] First, determine the main crops in the target area to be optimized, determine the total cultivated land area, total regional water supply, and regional population according to the regional annual plan, and determine the regional per capita food demand, water demand per unit area of ​​each crop, and yield per unit area of ​​each crop according to the regional historical yearbook. Determine the carbon absorption coefficient per unit yield of each crop, the carbon emission coefficient per unit area of ​​each crop, the use of each production factor per unit area of ​​each crop, and the carbon emission coefficient per unit use of each production factor according to the search engine. Determine the unit yield price of each crop and the fixed cost per unit area of ​​each crop according to local commodity trading data. The fixed costs include labor costs, seed costs, diesel costs, fertilizer costs, pesticide costs, and agricultural film costs. The production factors include diesel, fertilizers, pesticides, and agricultural films. Construct an agricultural production database for the target area to be optimized based on the above data information.

[0107] After constructing the agricultural production database of the area to be optimized, the regional agricultural planting structure optimization model is constructed by referring to the data information in the database. The water-saving benefit and economic benefit are considered at the same time for the constructed objective function. The water-saving benefit is expressed by the total water demand of crops and the difference between carbon absorption and carbon emissions to construct the agricultural planting structure optimization model.

[0108] In a specific embodiment, the expression of the objective function corresponding to the total amount of water required by the crops is:

[0109]

[0110] Among them, f 1 (x) represents the total water requirement of crops, w i represents the water requirement per unit area of ​​the i-th crop, x i represents the unit planting area of ​​the i-th crop, and I represents the total number of crop types;

[0111] The expression of the objective function corresponding to the economic benefit is:

[0112]

[0113] Among them, f 2 (x) represents economic benefit, p i represents the unit output price of the i-th crop, y i represents the yield per unit area of ​​the i-th crop, c i represents the fixed cost per unit area of ​​the i-th crop, which includes any number of labor cost, seed cost, diesel cost, fertilizer cost, pesticide cost and agricultural film cost, x i represents the unit planting area of ​​the i-th crop, and I represents the total number of crop types;

[0114] The expression of the objective function corresponding to the difference between the carbon absorption and the carbon emission is:

[0115]

[0116] Among them, f 3 (x) represents the difference between carbon absorption and carbon emission, a i represents the carbon absorption coefficient per unit yield of the i-th crop, y i represents the yield per unit area of ​​the i-th crop, e i represents the carbon emission coefficient per unit area of ​​the i-th crop, g j represents the emission coefficient per unit usage of the jth production factor, b ij represents the usage per unit area of ​​the jth production factor on the ith crop, where the production factor includes any number of diesel, fertilizer, pesticide and agricultural film, x i represents the unit planting area of ​​the i-th crop, I represents the total number of crop types, and J represents the total number of production factors;

[0117] The expression of the total cultivated land area constraint is:

[0118]

[0119] Among them, S represents the total area of ​​agricultural land in the target area, x i represents the unit planting area of ​​the i-th crop, and I represents the total number of crop types;

[0120] The expression of the regional water supply total amount constraint is:

[0121]

[0122] Where W represents the total water supply for agricultural cultivation in the target area, x i represents the unit planting area of ​​the i-th crop, and I represents the total number of crop types;

[0123] The expression of the food security constraint is:

[0124]

[0125] Where N represents the annual per capita food demand in the target area, K represents the population of the target area, and y i represents the yield per unit area of ​​the i-th crop, x i represents the unit planting area of ​​the i-th crop, and I represents the total number of crop types;

[0126] The expression for the non-negative constraint of the difference between carbon absorption and carbon emission is:

[0127]

[0128] Among them, a i represents the carbon absorption coefficient per unit area of ​​the i-th crop, y i represents the yield per unit area of ​​the i-th crop, e i represents the carbon emission coefficient per unit area of ​​the i-th crop, g j represents the emission coefficient per unit usage of the i-th production factor, b ij represents the usage per unit area of ​​the jth production factor in the ith crop, x i represents the unit planting area of ​​the i-th crop, I represents the total number of crop types, and J represents the total number of production factors.

[0129] It can be seen from the above embodiments that by adding the total water demand of crops as an optimization factor in the objective function, the optimization algorithm can effectively reduce the water resource consumption of agricultural irrigation in the region. Reasonable crop planting structure can improve the efficiency of water resource use and reduce over-irrigation, thereby achieving the purpose of water saving. The conservation of water resources not only helps to improve the resource utilization efficiency of agricultural production, but also reduces the excessive dependence on groundwater and other water sources, and avoids the over-exploitation of water resources. By optimizing the planting structure of crops, the algorithm can reasonably allocate the planting area of ​​different crops, thereby increasing the output value and farmers' income. The market price, yield, cost and other factors of crops are considered in the optimization process to ensure that the best economic return can be obtained under the constraints of regional resources. This can dynamically adjust the planting structure according to the market demand and supply and demand changes of agricultural production, and improve the overall benefits of the regional agricultural economy. The balance of carbon absorption and emission of crops, especially crops, is also considered in the optimization process. By optimizing the planting area of ​​different crops, the carbon absorption capacity can be improved and the emission of greenhouse gases can be reduced at the same time. The photosynthesis and carbon fixation of crops can effectively absorb carbon dioxide, and by reasonably optimizing the planting structure, carbon emissions in agricultural activities (such as the use of fertilizers and pesticides) can also be reduced.

[0130] Step S20, taking the minimum total water requirement of the crops, the maximum economic benefit and the minimum difference between the carbon absorption and the carbon emission as the optimization direction, using a preset first agricultural planting structure optimization algorithm to randomly generate a first population and a second population, iterating the first population and the second population to determine a first offspring population corresponding to the first population and a second offspring population corresponding to the second population, wherein the first agricultural planting structure optimization algorithm is a constrained multi-objective evolutionary algorithm based on a dual-population co-evolution strategy;

[0131] Step S30, determining a first mixed population and a second mixed population respectively according to the first population, the second population, the first offspring population and the second offspring population, and calculating and determining a first fitness value corresponding to each population individual in the first mixed population and a second fitness value corresponding to each population individual in the second mixed population respectively according to the objective function and the constraint conditions, wherein the population includes a plurality of population individuals, and each population individual represents a planting area corresponding to various crops;

[0132] Step S40, looping through the above steps until the difference between the sum of the second fitness values ​​in the second mixed population and the sum of the fitness values ​​of the previous generation second population is less than a preset threshold, and the individual solutions in the second population are all non-dominated solutions, outputting the individual solution set of the first population and the individual solution set of the second population;

[0133] After obtaining the decision variables, the objective function and the constraints, the total amount of water required by the crops is minimized, the economic benefit is maximized and the difference between the carbon absorption and the carbon emission is minimized as the optimization direction, a first population and a second population are randomly generated by using a preset first agricultural planting structure optimization algorithm, the first population and the second population are iterated to determine a first offspring population corresponding to the first population and a second offspring population corresponding to the second population, a first mixed population and a second mixed population are respectively determined according to the first population, the second population, the first offspring population and the second offspring population, and the first mixed population and the second mixed population are respectively calculated and determined according to the objective function and the constraints. The first fitness value corresponding to each population individual in the mixed population and the second fitness value corresponding to each population individual in the second mixed population are looped to execute the above steps until the difference between the sum of the second fitness values ​​in the second mixed population and the sum of the fitness values ​​of the previous generation of the second population is less than a preset threshold, and the individual solutions in the second population are all non-dominated solutions, then the individual solution set of the first population and the individual solution set of the second population are output, wherein the first agricultural planting structure optimization algorithm is a constrained multi-objective evolutionary algorithm based on a dual-population co-evolution strategy, the population includes a plurality of population individuals, and each population individual represents the planting area corresponding to various crops.

[0134] In some embodiments, see Figure 2 The steps of randomly generating a first population and a second population using a preset first agricultural planting structure optimization algorithm, and iterating the first population and the second population to determine a first offspring population corresponding to the first population and a second offspring population corresponding to the second population include:

[0135] Step S201, taking the minimum total water requirement of the crops, the maximum economic benefit and the minimum difference between the carbon absorption and carbon emission as the optimization direction, the first agricultural planting structure optimization algorithm is used to initialize the first population, the second population, the population size, the maximum number of iterations, the current number of iterations variable and the preset threshold, wherein the current number of iterations variable is t 1 , the population size is N, and the maximum number of iterations is T 1max， The preset threshold is α, which is used to determine whether the population falls into a stopped state;

[0136] Step S202: using a random selection strategy to generate a first mating pool according to the first population, and using simulated binary crossover and polynomial mutation to randomly select parent individuals from the first mating pool to generate offspring individuals, so as to determine the first offspring population;

[0137] Step S203, using a random selection strategy to generate a second mating pool according to the second population, and using simulated binary crossover and polynomial mutation to randomly select parent individuals from the second mating pool to generate offspring individuals, so as to determine the second offspring population;

[0138] Step S204: merging the first population, the first offspring population, and the second offspring population to determine the first mixed population, combining constraint dominance rule sorting with crowding distance sorting to calculate and determine the fitness value corresponding to the first mixed population, and selecting the best multiple population individuals as the first population of the next generation;

[0139] Step S205: merge the second population, the first offspring population and the second offspring population to determine the second mixed population, calculate and determine the fitness value corresponding to the second mixed population by combining non-dominated sorting and crowding distance sorting, and select the best multiple population individuals as the next generation second population;

[0140] Step S206, loop through the above steps until the difference between the sum of the second fitness values ​​in the second mixed population and the sum of the fitness values ​​of the previous generation second population is less than a preset threshold α, and the individual solutions in the second population are all non-dominated solutions, output the individual solution set of the first population and the individual solution set of the second population; or until the current iteration number t 1 Reach the maximum number of iterations T 1max , then stop the iteration and output the individual solution set of the first population and the individual solution set of the second population; otherwise, let t 1 =t 1 +1, return to the above step S202 to continue iterative evolution.

[0141] Specifically, the mating pool is a set of individuals used to reproduce the next generation in the genetic algorithm. The optimization direction is to minimize the total water requirement of the crops, maximize the economic benefits, and minimize the difference between the carbon absorption and carbon emissions. The planting area of ​​each crop is used as a decision variable for real number encoding. The first round of optimization is initialized, and the population size N and the maximum number of iterations T for the first round of optimization are set. 1max ; Randomly generate the first population and the second population, and record the sum of all target values ​​of all individuals in the initial second population as the sum of the initial overall target value, and set the current iteration number variable t 1 Set to 1 to set the threshold α used to determine whether the population falls into a stopped state;

[0142] Since the set constraints will hinder the evolution of the population, if the constraints create a large infeasible area in the decision space, it will cause the first population to be difficult to converge. Therefore, the second population is used to explore the target space without considering the constraints, and the offspring information is used to help the main population converge. The specific steps are as follows:

[0143] Randomly select an individual from the first population and enter the first mating pool, and repeat this step until the number of individuals in the first mating pool is N; randomly select two individuals from the first mating pool, and then perform simulated binary crossover and polynomial mutation operations on them to generate an offspring individual to enter the first offspring population, and repeat this step until the number of individuals in the first offspring population is N; randomly select an individual from the second population and enter the second mating pool, and repeat this step until the number of individuals in the second mating pool is N; randomly select two individuals from the second mating pool, and then perform simulated binary crossover and polynomial mutation operations on them to generate an offspring individual to enter the second offspring population, and repeat This step is continued until the number of the second offspring population is N; the first population, the first offspring population, and the second offspring population are combined to obtain a first mixed population, and the fitness values ​​of the individuals in the first mixed population are calculated by combining the constraint dominance rule sorting that strictly follows the constraint conditions with the crowding distance sorting, and the best N individuals are selected as the first population of the next generation according to the fitness sorting; the second population, the first offspring population, and the second offspring population are combined to obtain a second mixed population, and the fitness values ​​of the individuals in the second mixed population are calculated by combining the non-dominated sorting that completely ignores the constraint conditions with the crowding distance sorting, and the best N individuals are selected as the second population of the next generation;

[0144] Record the sum of the overall target values ​​of the second population at this time; then calculate the difference between the sum of the overall target values ​​of the second population of two generations. If the difference is less than the set threshold α and all the solutions in the second population are non-dominated solutions, it is determined that the evolution of the second population has stagnated. If the current number of iterations t 1 Equal to T 1maxOr if the population evolution is stagnant, the first round of iteration is stopped, and the individual solution set of the first population and the individual solution set of the second population are output as the first round optimization result; otherwise, let t 1 =t 1 +1, return to the above step S202 to continue iterative evolution.

[0145] Step S50: use the individual solution set of the second population as the individual solution set of the third population, and use a preset second agricultural planting structure optimization algorithm to iterate the individual solution set of the first population, the individual solution set of the second population, and the individual solution set of the third population. When the number of iterations reaches a preset number of iterations, output the non-dominated solutions in the first population as the optimal agricultural planting structure solution set, wherein the second agricultural planting structure optimization algorithm is a constrained multi-objective evolutionary algorithm based on a dual traction mechanism and a dynamic search attention strategy, and the dual traction mechanism includes a binary tournament selection strategy and a random selection strategy.

[0146] The agricultural planting structure scheme set represents the individual solution set constructed by the corresponding planting areas of various crops; after outputting the individual solution set of the first population and the individual solution set of the second population, the individual solution set of the second population is used as the individual solution set of the third population, and the individual solution set of the first population, the individual solution set of the second population and the individual solution set of the third population are iterated using a preset second agricultural planting structure optimization algorithm. When the number of iterations reaches a preset number of iterations, the non-dominated solution in the first population is output as the optimal agricultural planting structure scheme set, wherein the second agricultural planting structure optimization algorithm is a constrained multi-objective evolutionary algorithm based on a dual traction mechanism and a dynamic search attention strategy, and the dual traction mechanism includes a binary tournament selection strategy and a random selection strategy.

[0147] In some embodiments, see Figure 3 , taking the individual solution set of the second population as the individual solution set of the third population, using a preset second agricultural planting structure optimization algorithm to iterate the individual solution set of the first population, the individual solution set of the second population, and the individual solution set of the third population, when the number of iterations reaches a preset number of iterations, outputting the non-dominated solutions in the first population as the optimal agricultural planting structure solution set, comprising:

[0148] Step S501, taking the minimum total water requirement of the crops, the maximum economic benefit and the minimum difference between the carbon absorption and carbon emission as the optimization direction, taking the individual solution set of the second population as the individual solution set of the third population;

[0149] Step S502: Use the preset second agricultural planting structure optimization algorithm to initialize the ideal point, the maximum constraint violation value of the second population, the population size, the maximum number of iterations, the current number of iterations, and the minimum range of mating allowed between individuals, and generate a set of uniform reference vector sets in the target space, wherein the ideal point is Z * , which represents the minimum value of each objective value in the second population, and the maximum constraint violation value of the second population is ε 0 , the population size is N, and the maximum number of iterations is T 2max , the current iteration number variable is t 2 , the minimum range of mating allowed between individuals is β, and the reference vector set is Λ={λ 1 ,λ 2 ,λ 3 ,...,λ N};

[0150] Step S503, calling the preset double traction mechanism, using the binary tournament selection strategy to generate a third mating pool according to the first population, using simulated binary crossover and polynomial mutation to randomly select parent individuals from the third mating pool to generate offspring individuals, so as to determine the third offspring population;

[0151] Step S504: using a random selection strategy to generate a fourth mating pool according to the second population, and using simulated binary crossover and polynomial mutation to randomly select parent individuals from the fourth mating pool to generate offspring individuals, so as to determine a fourth offspring population;

[0152] Step S505: using a dynamic search attention strategy to select individuals entering the fifth mating pool and parent individuals participating in mating according to the third population, using simulated binary crossover and polynomial mutation to generate offspring individuals according to the parent individuals to determine the fifth offspring population;

[0153] Step S506: merge the first population, the third offspring population, the fourth offspring population and the fifth offspring population to determine a third mixed population, combine the constraint dominance rule sorting and the crowding distance sorting to calculate and determine the fitness value corresponding to the third mixed population, and select the best multiple population individuals as the first population of the next generation;

[0154] Step S507: merge the second population, the third offspring population, the fourth offspring population and the fifth offspring population to determine the fourth mixed population, combine the ε-constraint method with the crowding distance sorting calculation to determine the fitness value corresponding to the fourth mixed population, and select the best multiple population individuals as the second population of the next generation, wherein the calculation formula of the ε value used in this iteration in the ε-constraint method is expressed as:

[0155]

[0156] Among them, ε(t 2 ) represents the ε value of this iteration, t 2 Indicates the current iteration number, T 2max represents the maximum number of iterations, F 2 represents the feasible rate of the second population, ε 0 The maximum constraint violation value of the second population at the beginning of the second round of optimization;

[0157] Step S508, merging the third population, the third offspring population, the fourth offspring population and the fifth offspring population to obtain a fifth mixed population, and selecting a plurality of population individuals with low constraint violation values ​​and good diversity from the fifth mixed population as the third population of the next generation according to an environment selection strategy based on a reference vector and a constraint violation value;

[0158] Step S509: If the current number of iterations is t 2 Equal to the maximum number of iterations T 2max , stop the iteration, output the non-dominated solutions in the first population as the optimal agricultural planting structure solution set, otherwise, let t 2 =t 2 +1, repeat the above steps until the optimal agricultural planting structure solution set is output.

[0159] Specifically, the optimization direction is to minimize the total water requirement of the crops, maximize the economic benefits, and minimize the difference between the carbon absorption and carbon emissions, and the planting area of ​​each crop is used as the decision variable for real number encoding; the second round of optimization is initialized, the second population obtained in the first round of optimization is copied to the third population, the minimum value of each target value in the second population is calculated, and the ideal point Z is generated. * , calculate the maximum constraint violation value of the second population and record it as ε 0 , set the population size N and the maximum number of iterations T 2max , set the current iteration number variable t 2 Set to 1, set the minimum range β allowed for mating between individuals; generate a set of uniform reference vectors Λ={λ 1 ,λ 2 ,λ 3 ,...,λ N The reference vector set is obtained by using the normal boundary intersection method to generate a set of uniformly distributed reference points in the target space, and then connecting the coordinate origin with these reference points to obtain the reference vector set Λ={λ 1 ,λ 2 ,λ 3 ,...,λ N};

[0160] Although the first population after the first round of optimization can obtain a set of individual solutions with good convergence, when the unconstrained Pareto front is too far away from the constrained Pareto front, it is difficult for the second population to continue to provide useful information for the first population in terms of diversity, which may cause the first population to fall into the dilemma of local optimality and lack the motivation to jump out of the local optimality. Therefore, the second population uses the ε-constraint method to approach the constrained Pareto frontier to continue to provide convergence motivation for the main population. The third population uses the environment selection strategy based on the reference vector and the constraint violation value to obtain solutions with good diversity and low constraint violation values, providing motivation for the main population to improve diversity. The specific steps are as follows:

[0161] The fitness of individuals in the first population is calculated by combining the constraint dominance rule with the crowding distance sorting; two individuals are randomly selected from the first population, and the individuals with high fitness are selected to enter the third mating pool, and this step is repeated until the number of individuals in the third mating pool is N; two individuals are randomly selected from the third mating pool, and then simulated binary crossover and polynomial mutation operations are performed on them to generate an offspring individual to enter the third offspring population, and this step is repeated until the third offspring population is N; an individual is randomly selected from the second population to enter the fourth mating pool, and this step is repeated until the number of individuals in the fourth mating pool is N; two individuals are randomly selected from the fourth mating pool, and then simulated binary crossover and polynomial mutation operations are performed on them to generate an offspring individual to enter the fourth offspring population, and this step is repeated until the number of the fourth offspring population is N; in the third population, a dynamic search attention strategy is used to select individuals entering the fifth mating pool and parent individuals participating in mating, and then simulated binary crossover and polynomial mutation operations are performed on them to generate an offspring individual to enter the fifth offspring population, and this step is repeated. The steps are repeated until the number of the fifth offspring population is N; the first population, the third offspring population, the fourth offspring population and the fifth offspring population are merged to obtain a third mixed population, and the fitness of the individuals in the third mixed population is calculated by combining the constraint dominance rule sorting that strictly follows the constraint conditions with the crowding distance sorting, and the best N individuals are selected as the first population of the next generation according to the fitness sorting; the second population, the third offspring population, the fourth offspring population and the fifth offspring population are merged to obtain a fourth mixed population, and the ε value under the current number of iterations is calculated according to the calculation formula of the ε value in step S507, and the fitness is calculated by combining the ε-constraint method that relaxes the constraint conditions with the crowding distance sorting, and the best N individuals are selected as the second population of the next generation according to the fitness sorting; the third population, the third offspring population, the fourth offspring population and the fifth offspring population are merged to obtain a fifth mixed population, and N individuals with low constraint violation values ​​and good diversity are selected from the third mixed population by adopting the environmental selection strategy based on the reference vector and the constraint violation value as the third population of the next generation;

[0162] Determine whether to stop the iteration. If the current number of iterations is t 2Equal to the maximum number of iterations T 2max , stop the second round of optimization, output the non-dominated solutions in the first population as the second round of optimization results, to determine the optimal agricultural planting structure solution set, and output the second round of optimization results as the final agricultural planting structure solution set; otherwise, let t 2 =t 2 +1, return to the above step S503 to continue iterative evolution.

[0163] For further examples, see Figure 4 The steps of selecting individuals entering the fifth mating pool and parent individuals participating in mating according to the third population by using a dynamic search attention strategy, and generating offspring individuals according to the parent individuals by using simulated binary crossover and polynomial mutation to determine the fifth offspring population include:

[0164] Step S5051: Select N in the third group using the binary tournament selection strategy. fit individuals enter the fifth mating pool, where N fit The value calculation formula is expressed as:

[0165]

[0166] Among them, N fit represents the number of individuals selected in the third population using the binary tournament selection strategy, t 2 Indicates the current iteration number, T 2max Indicates the maximum number of iterations;

[0167] Step S5052: Select NN in the third group using a random selection strategy fit Individuals enter the fifth mating pool, and the range of mating allowed between two individuals at the current iteration number is calculated and determined, where the range of mating allowed between two individuals at the current iteration number is the Dis value;

[0168] Step S5053: When the Dis value is less than the minimum range β allowed for mating between two individuals, the Dis value is set equal to β. The calculation formula of the Dis value is expressed as:

[0169]

[0170] Among them, the Dis value represents the range of mating allowed between two individuals in the current iteration number, t 2 Indicates the current iteration number, T 2max Indicates the maximum number of iterations;

[0171] Step S5054: For each individual in the fifth mating pool, a parent individual is randomly selected from the Dis individuals in the fifth mating pool with the closest Euclidean distance to it to mate with it, and offspring individuals are generated by simulating binary crossover and polynomial mutation to form a fifth offspring population.

[0172] Specifically, in order to improve the optimization ability of the method, the third mating pool is mainly composed of randomly selected individuals in the early stage, because the influence of fitness is ignored, it has better diversity, is more suitable for global exploration, and improves the diversity of the population. In the later stage, individuals selected from the binary tournament are mainly used, because fitness is taken into account, which is suitable for local optimization and improves the quality of individuals; the mating range between individuals will affect the range that the offspring can search. A larger mating range between individuals in the early stage can be beneficial to global exploration, and a smaller mating range between individuals in the later stage is beneficial to local optimization; the above selection of the mating pool combined with the selection of the mating parent individual can improve the ability of the method to find an excellent solution set. The specific steps are as follows:

[0173] The fitness of the third population is calculated by combining non-dominated sorting and crowding distance sorting. According to step S5051, N is calculated. fit Randomly select two individuals from the third population, select the one with higher fitness to enter the fifth mating pool, and repeat this step until the number of individuals in the fifth mating pool is N fit ; Randomly select an individual from the third population to enter the fifth mating pool, repeat this step until the number of individuals in the fifth mating pool is N; calculate the Euclidean distance between each individual in the fifth mating pool; number the individuals in the fifth mating pool, take individual No. 1 as an example, individual No. 1 randomly selects an individual from the Dis individuals with the closest Euclidean distance in the fifth mating pool, and the two individuals use simulated binary crossover and polynomial mutation to generate an offspring individual to enter the fifth offspring population, traverse the fifth mating pool until each individual completes this step; finally generate a fifth offspring population of size N.

[0174] In a further embodiment, the step of merging the third population, the third offspring population, the fourth offspring population and the fifth offspring population to obtain a fifth mixed population, and selecting a plurality of population individuals with low constraint violation values ​​and good diversity from the fifth mixed population as the third population of the next generation according to an environment selection strategy based on a reference vector and a constraint violation value, comprises:

[0175] Step S5081: Update the ideal point Z according to the target information of the first population and the second population. * , assigning the individuals in the first population and the fifth mixed population to N sub-regions through the reference vector set;

[0176] Step S5082: record the individuals in the first population associated with the i-th sub-region as Mi , the individual associated with the ith sub-region in the fifth mixed population is denoted as P i ;

[0177] Step S5083: If M i is an empty set and P i is an empty set, select the individual closest to the ith subregion from the first population to enter the third population of the next generation; otherwise, if M i is an empty set and P i is not an empty set, from P i Select the individual with the smallest constraint violation value, and then select g from these individuals te The individual with the smallest value enters the third population of the next generation; otherwise, i Select the individual with the smallest constraint violation value, and then select g from these individuals te The individual with the smallest value enters the third population of the next generation, where g te The value is the function value calculated by the modified Chebyshev method, and its calculation formula is expressed as:

[0178]

[0179] Among them, g te (x|λ,Z * ) represents the g of an individual te Value, λ h represents the hth target of the reference vector, f h (x) represents the h-th fitness value of an individual, Represents the hth fitness value of the ideal point.

[0180] Specifically, the ideal point Z is updated by using the characteristic that the second population continuously converges to the constraint boundary * , the third population can be gradually shrunk to the vicinity of the constraint boundary when distributed according to the reference vector, and individuals with good diversity and constraint violation values ​​are selected from the first population and the third mixed population to provide motivation for the first population to improve diversity. The specific steps are as follows:

[0181] Calculate the minimum value of each target of the first population and the second population, and update the ideal point Z * ; Subtract the ideal point Z from the individual target values ​​in the first population and the fifth mixed population * Then calculate the vector angle with all reference vectors; divide the sub-regions according to the number of reference vectors, and each individual is associated with the sub-region corresponding to the reference vector with the smallest vector angle; the individual associated with the i-th sub-region in the first population is recorded as M i , the individual associated with the ith sub-region in the fifth mixed population is denoted as P i ; if M i is an empty set and Pi is an empty set, select the individual with the smallest vector angle with the i-th reference vector from the first population to enter the third population of the next generation; otherwise, if M i is an empty set and P i is not an empty set, from P i Select the individuals with the smallest constraint violation value, and calculate g for these individuals according to step S40463 te value, and then select g from these individuals te The individual with the smallest value enters the third population of the next generation; otherwise, i Select the individuals with the smallest constraint violation value, and calculate g for these individuals according to step S5083 te value, and then select g from these individuals te The individual with the smallest value enters the third population of the next generation.

[0182] It can be seen from the above embodiments that, compared with the prior art, the present application aims at the problem that in the prior art, multiple objectives and constraints need to be considered when optimizing agricultural planting structure. Multiple objectives are often conflicting with each other. Pursuing one of the objectives often leads to the deterioration of other objectives. The existence of constraints will make the constrained multi-objective evolutionary algorithm easy to fall into the dilemma of local optimum or poor convergence during the optimization process, affecting the quality of the final optimization solution set. The present application includes but is not limited to the following beneficial effects:

[0183] First, the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm of the present application adopts a dual-population co-evolution strategy in the first round of optimization. This strategy introduces two groups of populations, which evolve separately and influence each other, and can avoid the problem of poor convergence caused by difficulty in crossing the infeasible area during the solution process. Dual-population co-evolution can enhance the diversity of the population and avoid the optimization process from converging to an undesirable solution too early. Therefore, the first round of optimization can effectively cross a large range of infeasible areas, provide a wider search space, and ensure that a more reasonable planting plan is found.

[0184] Secondly, the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm of this application adopts a dual traction mechanism and a dynamic search attention strategy in the second round of optimization. The dual traction mechanism includes a binary tournament selection strategy and a random selection strategy, which helps the algorithm avoid the dilemma of premature convergence and local optimal solution. Specifically, the binary tournament selection strategy ensures the competitiveness between individual solutions and ensures that only high-quality solutions can be selected, while the random selection strategy avoids the population from falling into the limitation of a single choice.

[0185] Thirdly, the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm of this application has enhanced the adaptive ability of the population in the search space by introducing a dynamic search attention strategy. In the early stage of optimization, the algorithm can conduct extensive exploration to find possible optimal solutions; and as the number of iterations increases, the attention gradually turns to local development and focuses more on the nearby areas of potential optimal solutions, thereby accelerating convergence and avoiding falling into local optimality.

[0186] Fourthly, the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm of the present application, through two rounds of optimization, the obtained agricultural planting structure solution set not only has better convergence, that is, it can quickly approach the optimal solution, but also has higher diversity and can provide multiple effective solutions. This diversity means that different agricultural planting plans can be flexibly selected according to different regional needs, climatic conditions and market conditions, so that they have stronger adaptability in practical applications.

[0187] Fifth, the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm of this application, by comprehensively considering multiple factors such as water resource conservation and economic benefits, the obtained agricultural planting structure optimization plan has better sustainable development potential. Through this optimization method, not only can the agricultural production efficiency and income be improved, but also it can ensure that agricultural activities can continue under the background of limited resources and increasing environmental pressure.

[0188] Furthermore, the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm of the present application provides a new solution for the optimization of regional agricultural planting structure, which has significant beneficial effects. Through reasonable algorithm design and multi-objective optimization, the agricultural planting structure can achieve a balance in ensuring water conservation and improving economic benefits. Combined with the dual-population co-evolution strategy and dual traction mechanism, the algorithm can improve convergence and avoid local optimal solutions, thereby improving the quality and practicality of the optimization solution.

[0189] See also Figure 5, a regional agricultural planting structure optimization device based on a constrained multi-objective evolutionary algorithm is provided to meet one of the purposes of the present application, including a parameter acquisition module 1100, a first planting structure optimization module 1200, a fitness value determination module 1300, an individual solution set determination module 1400 and a second planting structure optimization module 1500. Among them, the parameter acquisition module 1100 is configured to obtain decision variables, objective functions and constraints in response to instructions for optimizing the agricultural planting structure of the target area, wherein the decision variables are the planting areas corresponding to various crops, the objective function includes the total water demand of various crops corresponding to various crops, the economic benefits and the difference between carbon absorption and carbon emissions, the constraints include the total cultivated land area constraint, the regional water supply total amount constraint, the food security constraint and the difference between carbon absorption and carbon emissions is a non-negative constraint; the first planting structure optimization module 1200 is configured to minimize the total water demand of the crops, maximize the economic benefits and the carbon absorption The difference between the carbon emissions is a non-negative constraint. The minimum difference between carbon absorption and carbon emission is taken as the optimization direction, a first population and a second population are randomly generated by a preset first agricultural planting structure optimization algorithm, and the first population and the second population are iterated to determine a first offspring population corresponding to the first population and a second offspring population corresponding to the second population, wherein the first agricultural planting structure optimization algorithm is a constrained multi-objective evolutionary algorithm based on a dual-population co-evolution strategy; a fitness value determination module 1300 is configured to determine a first mixed population and a second mixed population according to the first population, the second population, the first offspring population and the second offspring population, respectively, according to the objective function The constraint conditions are respectively calculated to determine the first fitness value corresponding to each individual in the first mixed population and the second fitness value corresponding to each individual in the second mixed population, wherein the population includes a plurality of individuals, and each individual represents the planting area corresponding to various crops; the individual solution set determination module 1400 is configured to execute the above steps in a loop until the difference between the sum of the second fitness values ​​in the second mixed population and the sum of the fitness values ​​of the second population of the previous generation is less than a preset threshold, and the individual solutions in the second population are all non-dominated solutions, outputting the individual solution set of the first population and the second population. The second planting structure optimization module 1500 is configured to use the individual solution set of the second population as the individual solution set of the third population, and adopt a preset second agricultural planting structure optimization algorithm to iterate the individual solution set of the first population, the individual solution set of the second population and the individual solution set of the third population, and when the number of iterations reaches a preset number of iterations, output the non-dominated solutions in the first population as the optimal agricultural planting structure solution set, wherein the second agricultural planting structure optimization algorithm is a constrained multi-objective evolutionary algorithm based on a dual traction mechanism and a dynamic search attention strategy, and the dual traction mechanism includes a binary tournament selection strategy and a random selection strategy.

[0190] Based on any embodiment of this application, please refer to Figure 6 Another embodiment of the present application further provides an electronic device, which can be implemented by a computer device, such as Figure 6 As shown, a schematic diagram of the internal structure of a computer device. The computer device includes a processor, a computer-readable storage medium, a memory, and a network interface connected via a system bus. Among them, the computer-readable storage medium of the computer device stores an operating system, a database, and computer-readable instructions. The database may store a control information sequence. When the computer-readable instructions are executed by the processor, the processor can implement a regional agricultural planting structure optimization method based on a constrained multi-objective evolutionary algorithm. The processor of the computer device is used to provide computing and control capabilities to support the operation of the entire computer device. The memory of the computer device may store computer-readable instructions. When the computer-readable instructions are executed by the processor, the processor can execute the regional agricultural planting structure optimization method based on a constrained multi-objective evolutionary algorithm of the present application. The network interface of the computer device is used to connect and communicate with a terminal. Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0191] In this embodiment, the processor is used to execute Figure 5 The memory stores the program code and various data required to execute the above modules. The network interface is used to transmit data between user terminals or servers. The memory in this embodiment stores the program code and data required to execute all modules / submodules in the regional agricultural planting structure optimization device based on constrained multi-objective evolutionary algorithm of this application, and the server can call the program code and data of the server to execute the functions of all submodules.

[0192] The present application also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm described in any embodiment of the present application.

[0193] The present application also provides a computer program product, including a computer program / instruction, which, when executed by one or more processors, implements the steps of the regional agricultural planting structure optimization method based on a constrained multi-objective evolutionary algorithm as described in any embodiment of the present application.

[0194] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments of the present application can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a computer-readable storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0195] The above description is only a partial implementation method of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

[0196] In summary, the regional agricultural planting structure optimization method based on the constrained multi-objective evolutionary algorithm of this application provides a new solution for the optimization of regional agricultural planting structure, which has significant beneficial effects. Through reasonable algorithm design and multi-objective optimization, the agricultural planting structure can achieve a balance in ensuring water conservation and improving economic benefits. Combined with the dual-population co-evolution strategy and dual traction mechanism, the algorithm can improve convergence and avoid local optimal solutions, thereby improving the quality and practicality of the optimization solution.

Claims

1. A method for optimizing regional agricultural planting structure based on a constrained multi-objective evolutionary algorithm, characterized in that: include: In response to an instruction to optimize the agricultural planting structure of a target area, a decision variable, an objective function, and a constraint condition are obtained, wherein the decision variable is the planting area corresponding to various crops, the objective function includes the total water demand of crops corresponding to various crops, the economic benefit, and the difference between carbon absorption and carbon emission, and the constraint condition includes a constraint on the total cultivated land area, a constraint on the total regional water supply, a constraint on food security, and a non-negative constraint on the difference between carbon absorption and carbon emission; Taking the minimum total water requirement of the crops, the maximum economic benefit and the minimum difference between the carbon absorption and the carbon emission as the optimization direction, a preset first agricultural planting structure optimization algorithm is used to randomly generate a first population and a second population, and the first population and the second population are iterated to determine a first offspring population corresponding to the first population and a second offspring population corresponding to the second population, wherein the first agricultural planting structure optimization algorithm is a constrained multi-objective evolutionary algorithm based on a dual-population co-evolution strategy; Determine a first mixed population and a second mixed population according to the first population, the second population, the first offspring population, and the second offspring population, respectively, and calculate and determine a first fitness value corresponding to each population individual in the first mixed population and a second fitness value corresponding to each population individual in the second mixed population according to the objective function and the constraint conditions, wherein the population includes a plurality of population individuals, and each population individual represents a planting area corresponding to various crops; The above steps are executed repeatedly until the difference between the sum of the second fitness values ​​in the second mixed population and the sum of the fitness values ​​of the second population of the previous generation is less than a preset threshold, and the individual solutions in the second population are all non-dominated solutions, and the individual solution set of the first population and the individual solution set of the second population are output; The individual solution set of the second population is used as the individual solution set of the third population, and the preset second agricultural planting structure optimization algorithm is used to iterate the individual solution set of the first population, the individual solution set of the second population and the individual solution set of the third population. When the number of iterations reaches the preset number of iterations, the non-dominated solutions in the first population are output as the optimal agricultural planting structure solution set, wherein the second agricultural planting structure optimization algorithm is a constrained multi-objective evolutionary algorithm based on a dual traction mechanism and a dynamic search attention strategy, and the dual traction mechanism includes a binary tournament selection strategy and a random selection strategy.

2. The method for optimizing regional agricultural planting structure based on a constrained multi-objective evolutionary algorithm according to claim 1 is characterized in that: The expression of the objective function corresponding to the total amount of water required by crops is: Among them, f1(x) represents the total water requirement of crops, w i represents the water requirement per unit area of ​​the i-th crop, x i represents the unit planting area of ​​the i-th crop, and I represents the total number of crop types; The expression of the objective function corresponding to the economic benefit is: Among them, f2(x) represents economic benefits, p i represents the unit output price of the i-th crop, y i represents the yield per unit area of ​​the i-th crop, c i represents the fixed cost per unit area of ​​the i-th crop, which includes any number of labor cost, seed cost, diesel cost, fertilizer cost, pesticide cost and agricultural film cost, x i represents the unit planting area of ​​the i-th crop, and I represents the total number of crop types; The expression of the objective function corresponding to the difference between the carbon absorption and the carbon emission is: Where f3(x) represents the difference between carbon absorption and carbon emission, a i represents the carbon absorption coefficient per unit yield of the i-th crop, y i represents the yield per unit area of ​​the i-th crop, e i represents the carbon emission coefficient per unit area of ​​the i-th crop, g j represents the emission coefficient per unit usage of the jth production factor, b ij represents the usage per unit area of ​​the jth production factor on the ith crop, where the production factor includes any number of diesel, fertilizer, pesticide and agricultural film, x i represents the unit planting area of ​​the i-th crop, I represents the total number of crop types, and J represents the total number of production factors; The expression of the total cultivated land area constraint is: Among them, S represents the total area of ​​agricultural land in the target area, x i represents the unit planting area of ​​the i-th crop, and I represents the total number of crop types; The expression of the regional water supply total amount constraint is: Where W represents the total water supply for agricultural cultivation in the target area, x i represents the unit planting area of ​​the i-th crop, and I represents the total number of crop types; The expression of the food security constraint is: Where N represents the annual per capita food demand in the target area, K represents the population of the target area, and y i represents the yield per unit area of ​​the i-th crop, x i represents the unit planting area of ​​the i-th crop, and I represents the total number of crop types; The expression for the non-negative constraint of the difference between carbon absorption and carbon emission is: Among them, a i represents the carbon absorption coefficient per unit area of ​​the i-th crop, y i represents the yield per unit area of ​​the i-th crop, e i represents the carbon emission coefficient per unit area of ​​the i-th crop, g j represents the emission coefficient per unit usage of the i-th production factor, b ij represents the usage per unit area of ​​the jth production factor in the ith crop, x i represents the unit planting area of ​​the i-th crop, I represents the total number of crop types, and J represents the total number of production factors.

3. The method for optimizing regional agricultural planting structure based on a constrained multi-objective evolutionary algorithm according to claim 1 is characterized in that: The steps of randomly generating a first population and a second population by using a preset first agricultural planting structure optimization algorithm, and iterating the first population and the second population to determine a first offspring population corresponding to the first population and a second offspring population corresponding to the second population include: Taking the minimum total water requirement of the crops, the maximum economic benefit and the minimum difference between the carbon absorption and carbon emission as the optimization direction, the first agricultural planting structure optimization algorithm is used to initialize the first population, the second population, the population size, the maximum number of iterations, the current number of iterations variable and the preset threshold, wherein the current number of iterations variable is t1, the population size is N, and the maximum number of iterations is T 1max， The preset threshold is α, which is used to determine whether the population falls into a stopped state; Using a random selection strategy to generate a first mating pool according to the first population, and using simulated binary crossover and polynomial mutation to randomly select parent individuals from the first mating pool to generate offspring individuals, so as to determine the first offspring population; Using a random selection strategy to generate a second mating pool according to the second population, and using simulated binary crossover and polynomial mutation to randomly select parent individuals from the second mating pool to generate offspring individuals, so as to determine the second offspring population; Merging the first population, the first offspring population and the second offspring population to determine the first mixed population, combining constraint dominance rule sorting with crowding distance sorting to calculate and determine the fitness value corresponding to the first mixed population, and selecting the best multiple population individuals as the first population of the next generation; Merging the second population, the first offspring population and the second offspring population to determine the second mixed population, combining non-dominated sorting with crowding distance sorting to calculate and determine the fitness value corresponding to the second mixed population, and selecting the best multiple population individuals as the next generation second population; The above steps are executed repeatedly until the difference between the sum of the second fitness values ​​in the second mixed population and the sum of the fitness values ​​of the previous generation of the second population is less than a preset threshold α, and the individual solutions in the second population are all non-dominated solutions, then the individual solution set of the first population and the individual solution set of the second population are output; Or until the current number of iterations t1 reaches the maximum number of iterations T 1max , then stop the iteration and output the individual solution set of the first population and the individual solution set of the second population.

4. The method for optimizing regional agricultural planting structure based on a constrained multi-objective evolutionary algorithm according to claim 1, characterized in that: The step of using the individual solution set of the second population as the individual solution set of the third population, using a preset second agricultural planting structure optimization algorithm to iterate the individual solution set of the first population, the individual solution set of the second population, and the individual solution set of the third population, and when the number of iterations reaches a preset number of iterations, outputting the non-dominated solutions in the first population as the optimal agricultural planting structure solution set includes: Taking the minimization of the total amount of water required by the crops, the maximization of the economic benefits and the minimization of the difference between the carbon absorption and the carbon emission as the optimization direction, the individual solution set of the second population is used as the individual solution set of the third population; The preset second agricultural planting structure optimization algorithm is used to initialize the ideal point, the maximum constraint violation value of the second population, the population size, the maximum number of iterations, the current number of iterations, and the minimum range of mating allowed between individuals, and a set of uniform reference vector sets are generated in the target space, where the ideal point is Z * , which represents the minimum value of each target value in the second population, the maximum constraint violation value of the second population is ε0, the population size is N, and the maximum number of iterations is T 2max , the current iteration number variable is t2, the minimum range of mating allowed between individuals is β, and the reference vector set is Λ={λ1,λ2,λ3,...,λ N }; Calling a preset double traction mechanism, using a binary tournament selection strategy to generate a third mating pool according to the first population, and using simulated binary crossover and polynomial mutation to randomly select parent individuals from the third mating pool to generate offspring individuals, so as to determine a third offspring population; Using a random selection strategy to generate a fourth mating pool according to the second population, and using simulated binary crossover and polynomial mutation to randomly select parent individuals from the fourth mating pool to generate offspring individuals, so as to determine a fourth offspring population; A dynamic search attention strategy is used to select individuals entering the fifth mating pool and parent individuals participating in mating according to the third population, and simulated binary crossover and polynomial mutation are used to generate offspring individuals according to the parent individuals to determine the fifth offspring population; Merge the first population, the third offspring population, the fourth offspring population and the fifth offspring population to determine a third mixed population, combine the constraint dominance rule sorting and the crowding distance sorting to calculate and determine the fitness value corresponding to the third mixed population, and select the best multiple population individuals as the first population of the next generation; The second population, the third offspring population, the fourth offspring population and the fifth offspring population are combined to determine the fourth mixed population, and the fitness value corresponding to the fourth mixed population is determined by combining the ε-constraint method with the crowding distance sorting calculation, and the optimal multiple population individuals are selected as the second population of the next generation, wherein the calculation formula of the ε value used in this iteration in the ε-constraint method is expressed as: Among them, ε(t2) represents the ε value of this iteration, t2 represents the current iteration number, T 2max represents the maximum number of iterations, F2 represents the feasible rate of the second population, ε0 represents the maximum constraint violation value of the second population at the beginning of the second round of optimization; Merging the third population, the third offspring population, the fourth offspring population and the fifth offspring population to obtain a fifth mixed population, and selecting a plurality of population individuals with low constraint violation values ​​and good diversity from the fifth mixed population as the third population of the next generation according to an environmental selection strategy based on a reference vector and a constraint violation value; If the current number of iterations t2 is equal to the maximum number of iterations T 2max , stop the iteration, and output the non-dominated solutions in the first population as the optimal agricultural planting structure solution set.

5. The method for optimizing regional agricultural planting structure based on a constrained multi-objective evolutionary algorithm according to claim 4 is characterized in that: The steps of selecting individuals entering the fifth mating pool and parent individuals participating in mating according to the third population by adopting a dynamic search attention strategy, and generating offspring individuals according to the parent individuals by adopting simulated binary crossover and polynomial mutation to determine the fifth offspring population include: Select N in the third population using a binary tournament selection strategy fit individuals enter the fifth mating pool, where N fit The value calculation formula is expressed as: Among them, N fit represents the number of individuals selected in the third population using the binary tournament selection strategy, t2 represents the current number of iterations, and T 2max Indicates the maximum number of iterations; Use random selection strategy to select NN in the third population fit Individuals enter the fifth mating pool, and the range of mating allowed between two individuals at the current iteration number is calculated and determined, where the range of mating allowed between two individuals at the current iteration number is the Dis value; When the Dis value is less than the minimum range β allowed for mating between two individuals, let the Dis value be equal to β. The calculation formula of the Dis value is expressed as: Among them, the Dis value represents the range of mating allowed between two individuals in the current iteration, t2 represents the current iteration number, and T 2max Indicates the maximum number of iterations; Each individual in the fifth mating pool randomly selects a parent individual to mate with from the Dis individuals in the fifth mating pool with the closest Euclidean distance to it, and generates offspring individuals by using simulated binary crossover and polynomial mutation to form a fifth offspring population.

6. The method for optimizing regional agricultural planting structure based on a constrained multi-objective evolutionary algorithm according to claim 4 is characterized in that: The steps of merging the third population, the third offspring population, the fourth offspring population and the fifth offspring population to obtain a fifth mixed population, and selecting a plurality of population individuals with low constraint violation values ​​and good diversity from the fifth mixed population as the third population of the next generation according to an environmental selection strategy based on a reference vector and a constraint violation value, include: Update the ideal point Z according to the target information of the first population and the second population * , assigning the individuals in the first population and the fifth mixed population to N sub-regions through the reference vector set; The individual associated with the ith subregion in the first population is denoted as M i , the individual associated with the ith sub-region in the fifth mixed population is denoted as P i ; If M i is an empty set and P i is an empty set, select the individual closest to the ith subregion from the first population to enter the third population of the next generation; otherwise, if M i is an empty set and P i is not an empty set, from P i Select the individual with the smallest constraint violation value, and then select g from these individuals te The individual with the smallest value enters the third population of the next generation; otherwise, i Select the individual with the smallest constraint violation value, and then select g from these individuals te The individual with the smallest value enters the third population of the next generation, where g te The value is the function value calculated by the modified Chebyshev method, and its calculation formula is expressed as: Among them, g te (x|λ,Z * ) represents the g of an individual te Value, λ h represents the hth target of the reference vector, f h (x) represents the h-th fitness value of an individual, Represents the hth fitness value of the ideal point.

7. The method for optimizing regional agricultural planting structure based on a constrained multi-objective evolutionary algorithm according to any one of claims 1 to 6, characterized in that: The agricultural planting structure scheme set represents an individual solution set constructed by the corresponding planting areas of various crops; the crops include rice crops, wheat crops and cotton crops.

8. A regional agricultural planting structure optimization device based on a constrained multi-objective evolutionary algorithm, characterized in that: include: a parameter acquisition module, configured to acquire decision variables, objective functions, and constraints in response to an instruction to optimize the agricultural planting structure of a target area, wherein the decision variables are the planting areas corresponding to various crops, the objective functions include the total water demand of crops corresponding to various crops, economic benefits, and the difference between carbon absorption and carbon emissions, and the constraints include the total cultivated land area constraint, the total regional water supply constraint, the food security constraint, and the difference between carbon absorption and carbon emissions is a non-negative constraint; The first planting structure optimization module is configured to minimize the total water requirement of the crops, maximize the economic benefits, and minimize the difference between the carbon absorption and carbon emissions as the optimization direction, randomly generate a first population and a second population using a preset first agricultural planting structure optimization algorithm, iterate the first population and the second population to determine a first offspring population corresponding to the first population and a second offspring population corresponding to the second population, wherein the first agricultural planting structure optimization algorithm is a constrained multi-objective evolutionary algorithm based on a dual-population co-evolutionary strategy; A fitness value determination module is configured to respectively determine a first mixed population and a second mixed population according to the first population, the second population, the first offspring population, and the second offspring population, and respectively calculate and determine a first fitness value corresponding to each population individual in the first mixed population and a second fitness value corresponding to each population individual in the second mixed population according to the objective function and the constraint conditions, wherein the population includes a plurality of population individuals, and each population individual represents a planting area corresponding to various crops; The individual solution set determination module is configured to execute the above steps cyclically until the difference between the sum of the second fitness values ​​in the second mixed population and the sum of the fitness values ​​of the previous generation second population is less than a preset threshold, and the individual solutions in the second population are all non-dominated solutions, and then output the individual solution set of the first population and the individual solution set of the second population; The second planting structure optimization module is configured to use the individual solution set of the second population as the individual solution set of the third population, and adopt a preset second agricultural planting structure optimization algorithm to iterate the individual solution set of the first population, the individual solution set of the second population, and the individual solution set of the third population. When the number of iterations reaches a preset number of iterations, the non-dominated solutions in the first population are output as the optimal agricultural planting structure solution set, wherein the second agricultural planting structure optimization algorithm is a constrained multi-objective evolutionary algorithm based on a dual traction mechanism and a dynamic search attention strategy, and the dual traction mechanism includes a binary tournament selection strategy and a random selection strategy.

9. An electronic device, comprising a central processing unit and a memory, characterized in that: The central processing unit is used to call and run the computer program stored in the memory to execute the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: It stores a computer program implemented according to the method described in any one of claims 1 to 7 in the form of computer-readable instructions, and when the computer program is called and executed by a computer, the steps included in the corresponding method are executed.

Images