A Multi-Objective Optimization-Based Method and System for Crab-Rice Ratio Decision Making

Abstract

This invention discloses a method and system for determining the crab-rice ratio based on multi-objective optimization, belonging to the field of agricultural intelligent decision-making technology. The method includes: collecting historical production data and environmental parameters of crab-rice co-cultivation; constructing rice yield prediction models and crab yield prediction models; constructing a multi-objective optimization model that includes maximizing comprehensive economic benefits, minimizing rice seedling damage risk, minimizing rice lodging risk, and minimizing feed costs; iteratively solving the multi-objective optimization model using a multi-objective optimization solution method to obtain an optimal crab-rice ratio that meets the constraints; and updating the rice and crab yield prediction models and the multi-objective optimization model based on actual production data after implementing the optimal crab-rice ratio. This invention can achieve the scientific configuration of crab-rice co-cultivation production parameters by comprehensively considering economic benefits and various production risks, thereby improving the stability and adaptability of the ratio scheme.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of information technology and agricultural intelligent decision-making, specifically relating to a crab-rice ratio decision-making method and system based on multi-objective optimization. Background Technology

[0002] Rice-crab co-cultivation is an agricultural production model that combines rice planting with crab farming, offering advantages in increasing overall yield per unit area and improving the farmland ecological environment. However, in actual production, rice-crab co-cultivation involves numerous key production parameters, including crab seedling stocking density, rice planting density, rice variety, crab size and stocking batch, as well as environmental conditions such as ditch ratio and water depth. Different combinations of parameters significantly impact rice yield, crab growth, and production risks.

[0003] In current production practices, the rice-crab ratio is usually determined based on manual experience or localized experimental results, lacking a systematic, calculable, and generalizable decision-making method. On the one hand, releasing an appropriate amount of river crabs can promote the ecological cycle of rice paddies to some extent; on the other hand, excessively high release densities of river crabs can easily damage rice seedlings, thereby increasing the risk of yield reduction and lodging. Furthermore, different regions exhibit significant differences in the characteristics of lodging and seedling damage over the years, making it difficult to comprehensively consider these various risk factors within the existing ratio decision-making process. This results in insufficient stability and adaptability of the ratio under different production environments.

[0004] Therefore, there is an urgent need for a technical method that can systematically model and optimize the production parameters of crab-rice co-cultivation under various constraints and risk objectives, so as to achieve the scientific configuration of crab-rice ratio schemes. Summary of the Invention

[0005] The purpose of this invention is to provide a crab-rice ratio decision-making method and system based on multi-objective optimization. By constructing a multi-objective optimization model that includes economic benefits and various production risks, and iteratively solving the model under constraints, a crab-rice ratio scheme that achieves an optimal or near-optimal balance between comprehensive benefits and risks is obtained, thereby improving the rationality, stability and replicability of crab-rice co-cultivation production decisions.

[0006] The first aspect of this invention provides a crab-rice ratio decision-making method based on multi-objective optimization, comprising the following steps:

[0007] Step S1: Collect historical crab-rice co-cultivation production data and environmental parameters; the historical crab-rice co-cultivation production data includes crab seedling release density. Rice planting density Rice varieties Initial specifications of river crabs River crab release batches River crab production and rice yield The environmental parameters include the proportion of ditches. Average water depth ,temperature and dissolved oxygen ;

[0008] Step S2: Preprocess the historical crab-rice co-cultivation production data and environmental parameters collected in Step S1;

[0009] Step S3: Based on the preprocessed historical rice-crab co-cultivation production data and environmental parameters, construct rice yield prediction models and crab yield prediction models respectively;

[0010] Step S4: Based on the output results of the rice yield prediction model and the crab yield prediction model, construct a multi-objective optimization model for the crab-rice ratio that includes the objectives of maximizing comprehensive economic benefits, minimizing feed costs, minimizing rice seedling damage risk, and minimizing rice lodging risk.

[0011] Step S5: Introduce constraints into the multi-objective optimization model and perform iterative solutions to obtain the optimal solution set that satisfies the constraints;

[0012] Step S6: Based on decision preferences, select the optimal solution from the set of optimal solutions as the crab-rice ratio optimization scheme and output it; the output results include: the optimized crab seedling release density. Rice planting density Corresponding rice varieties River crab specifications and batches of distribution And the values ​​of each objective function under the optimal solution.

[0013] Furthermore, in step S1, the historical crab-rice co-cultivation production data also includes average plant height. Feed costs and the risk of seedling damage and the risk of lodging Among them, seedling damage risk is the proportion of seedling loss area caused by crab activity to the planted area; lodging risk is the proportion of rice lodging area to the planted area; crab seedling release density is defined as the number of crabs released per unit area of ​​paddy field, in units of crabs / hm². 2 Rice planting density is defined as the number of rice plants per unit area of ​​paddy field, expressed as plants per hectare (hm²). 2 .

[0014] Furthermore, in step S2, the preprocessing of the historical crab-rice co-cultivation production data and environmental parameters collected in step S1 specifically includes:

[0015] The collected historical rice-crab co-cultivation production data underwent missing value processing, outlier removal, unit unification, standardization, and feature coding. Discrete variables such as rice variety, initial crab size, and release batch were encoded using one-hot or ordinal coding, while continuous variables such as ditch ratio, average water depth, temperature, dissolved oxygen, rice planting density, and crab release density were normalized or standardized to construct a numerical feature matrix. Simultaneously, rice yield, crab yield, feed cost, lodging risk, and seedling damage risk were used as target variables to construct corresponding label vectors.

[0016] Furthermore, in step S3, based on the pretreated rice yield... Corresponding crab seedling stocking density Rice planting density Rice varieties Environmental related parameters (Ditch ratio) ,temperature Average water depth Dissolved oxygen To construct a rice yield prediction model based on the relationship between the two factors:

[0017]

[0018] in, ~ The correlation coefficient is obtained by fitting historical crab-rice co-cultivation production data. This indicates the land occupation effect of ditches; This indicates that rice planting density is not necessarily better the higher it is, but rather there is an optimal value; This indicates that river crabs have a negative effect on rice. It reflects a cooperative and competitive relationship; This indicates that higher temperatures are not necessarily better; rather, there exists an optimal value. This indicates the positive effect of average water depth and dissolved oxygen.

[0019] Furthermore, in step S3, based on the pre-processed crab yield... With the density of crab seedlings Rice planting density Environmental related parameters (Ditch ratio) ,temperature Average water depth Dissolved oxygen Based on the relationship between ), a crab production prediction model is constructed:

[0020]

[0021] in, ~ The correlation coefficient is obtained by fitting historical crab-rice co-cultivation production data. This indicates that the stocking density of crab larvae is not necessarily better the higher it is, but rather there is an optimal value; This indicates that low-density rice has a positive effect on the growth of river crabs, while high-density rice has a negative effect due to spatial suppression. This indicates the existence of an optimal combination of ditch ratio, average water depth, and temperature. This indicates the positive effect of dissolved oxygen on the growth of crabs.

[0022] Furthermore, in step S4, the multi-objective optimization model includes the following objective function:

[0023] 1) Objective function for maximizing overall economic benefits

[0024] Based on the predicted rice and crab yields, and combined with the corresponding market price parameters, a comprehensive economic return function is constructed. And maximizing it is the primary optimization objective:

[0025]

[0026]

[0027] in, For rice profits; For the benefit of river crabs; For feed costs; The market price for rice is (yuan / kg). Rice yield (kg / hm) 2 ); The market price for river crabs is (yuan / kg). River crab yield (kg / hm) 2 ); The unit price of feed (yuan / kg); To match the initial size of the river crab and batches of river crabs released The relevant average weight gain, ; To match the specifications of river crabs and batches of river crabs released The relevant feed conversion ratio, ; Survival rate; Obtained by fitting historical crab-rice co-cultivation production data;

[0028] 2) Objective function for minimizing feed costs

[0029] Based on the initial size of the crabs, the batch size, and the corresponding feed consumption level, a feed cost function is constructed, and its minimization is used as the second optimization objective:

[0030]

[0031] in, The unit price of feed (yuan / kg); To match the specifications of river crabs and batches of river crabs released The relevant average weight gain, ; To match the specifications of river crabs and batches of river crabs released The relevant feed conversion ratio, ; Survival rate; Obtained by fitting historical crab-rice co-cultivation production data;

[0032] 3) Objective function for minimizing the risk of rice seedling damage

[0033] Based on the release density of crab seedlings, the planting density of rice, and environmental variables, a function characterizing the risk of damage to rice seedlings caused by crab activity is constructed, and its minimization is used as the third optimization objective.

[0034]

[0035] in, The average water depth is (m). The proportion of the ditch area, Dissolved oxygen (mg / L) ~ These are the fitting coefficients obtained by fitting historical crab-rice co-cultivation production data; This indicates that an excessive number of crabs leads to increased damage to seedlings; This indicates that dense rice paddies result in strong resistance to damage. and This indicates that the depth of the water and the presence of numerous ditches increase crab activity, thereby increasing the risk of damaging the seedlings. High dissolved oxygen levels lead to stable crab activity, thus reducing the risk of injury to crab seedlings.

[0036] 4) Objective function for minimizing the risk of rice lodging

[0037] Based on rice planting density, rice variety information, and historical lodging data, a rice lodging risk function is constructed, and its minimization is used as the fourth optimization objective.

[0038]

[0039] in, This represents the average plant height. ~ These are the fitting coefficients obtained by fitting historical data; This indicates that the higher the planting density of rice, the more prone it is to lodging; This indicates that the plant is too tall and prone to falling over. This indicates that the higher the rice variety value, the stronger its resistance to lodging; This indicates that the water depth caused the root system to become unstable; This indicates high dissolved oxygen levels, healthy root system, and strong resistance to lodging.

[0040] The aforementioned comprehensive economic benefit objective function, seedling damage risk function, rice lodging risk function, and feed cost function together constitute the objective function vector:

[0041]

[0042] The decision variable vector is: .

[0043] Furthermore, in step S5, the constraint conditions include at least one of the following conditions:

[0044] Crab seedling stocking density With rice planting density The range of values ​​is constrained;

[0045] Ditch ratio Average water depth With dissolved oxygen Physical constraints;

[0046] Risk threshold constraints were set based on the occurrence of lodging and seedling damage over the years.

[0047] Furthermore, in step S5, the solution method includes: combinatorial optimization algorithm, multi-objective evolutionary algorithm, mathematical programming method, heuristic search algorithm, or any combination thereof.

[0048] Furthermore, the crab-rice ratio decision method also includes step S7: after the crab-rice ratio optimization scheme obtained in S6 is implemented, production data and result data in the actual production process are collected and entered into the historical crab-rice co-cultivation production database; based on the updated historical crab-rice co-cultivation production data, the relevant parameters of the rice yield prediction model, the crab yield prediction model and the multi-objective optimization model are updated to improve the convergence efficiency and solution stability of the multi-objective optimization solution process under different environmental conditions, thereby realizing the continuous optimization of the crab-rice ratio decision method.

[0049] A second aspect of the present invention provides a crab-rice ratio decision system based on multi-objective optimization, comprising:

[0050] The data acquisition and storage module is used to collect and store historical crab-rice co-cultivation production data and environmental parameters;

[0051] The yield prediction module is used to construct rice yield prediction models and crab yield prediction models based on the historical rice-crab co-cultivation production data and environmental parameters.

[0052] The multi-objective optimization modeling module is used to construct a multi-objective optimization model that includes maximizing comprehensive economic benefits, minimizing the risk of rice seedling damage, and minimizing the risk of rice lodging.

[0053] The optimization solution module is used to iteratively solve the multi-objective optimization model and output the optimal solution set that satisfies the constraints and the corresponding objective function values ​​under the solution.

[0054] The feedback update module is used to update historical rice-crab co-cultivation production data, rice and crab yield prediction models, and multi-objective optimization models based on the actual production data of the optimal solution.

[0055] The beneficial effects of this invention are:

[0056] (1) This invention has the ability to perform multi-objective collaborative optimization, while taking into account the benefits, the risk of seedling damage and the risk of lodging;

[0057] (2) This invention integrates crab-rice interaction, environmental factors and their nonlinear relationships through ecological coupling modeling;

[0058] (3) This invention can be directly applied to production practice, providing information on planting density, variety selection, and breeding strategies. Attached Figure Description

[0059] Figure 1 This is an overall flowchart of the crab-rice ratio decision-making method based on multi-objective optimization in an embodiment of the present invention.

[0060] Figure 2 This is a schematic diagram of the crab-rice ratio decision system based on multi-objective optimization in an embodiment of the present invention.

[0061] Figure 3 This is an example diagram of Pareto front for multi-objective optimization in an embodiment of the present invention. Detailed Implementation

[0062] To make the technical solution, technical features, and beneficial effects of the present invention clearer, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Those skilled in the art should understand that the following embodiments are for illustrative purposes only and do not constitute a limitation on the scope of protection of the present invention.

[0063] like Figure 1 As shown, this embodiment of the invention provides a crab-rice ratio decision-making method based on multi-objective optimization, including the following steps:

[0064] Step 1: Collect basic data, including historical crab-rice co-cultivation production data and environmental parameters from step S1 above.

[0065] Step 2, select an area of ​​10 hm² 2 For standardized plots, the following variables are defined:

[0066] Crab seedling stocking density: only / hm 2 ;

[0067] Rice planting density: , plant / hm 2 ;

[0068] River crab specifications: ;

[0069] Distribution batches: ;

[0070] Rice varieties: ;

[0071] Mapping relationship of lodging resistance coefficient of rice varieties: ;

[0072] Average plant height: cm;

[0073] The environmental parameters include:

[0074] Ditch ratio: ;

[0075] Average water depth: m;

[0076] Water temperature: ;

[0077] Dissolved oxygen: mg / L.

[0078] Step 3: Based on historical crab-rice co-cultivation production data and environmental parameters, a yield prediction model for crabs and rice is obtained.

[0079] 1) Crab production model

[0080]

[0081] 2) Rice yield model

[0082]

[0083] Step four: Construct a multi-objective optimization model for the crab-rice ratio, wherein the multi-objective optimization model includes the following objective function:

[0084] 1) Feed cost prediction model

[0085]

[0086]

[0087]

[0088] Comprehensive economic return function

[0089] Let the market price be: River Crab Price =78 yuan / kg; Rice price = 2.6 yuan / kg

[0090] The overall economic return function is:

[0091]

[0092] 3) Risk function of river crabs damaging rice seedlings

[0093]

[0094] 4) Rice lodging risk function

[0095]

[0096] A multi-objective optimization model is constructed to maximize overall economic benefits, minimize the risk of seedling damage, and minimize the risk of lodging.

[0097]

[0098] Step 5: Pareto optimization to solve the multi-objective optimization model.

[0099] The constraint is the crab larvae stocking density set in step two. With rice planting density The range of values ​​is constrained;

[0100] The multi-objective evolutionary algorithm is used to solve the above multi-objective optimization model to obtain the Pareto optimal solution set, such as... Figure 3 As shown in Table 1, three Pareto optimal solution sets can be selected.

[0101] Table 1. Three Pareto optimal solution sets obtained from solving the multi-objective optimization model.

[0102]

[0103] Step Six: Comparison and Verification

[0104] The schemes corresponding to maximizing single-objective benefits were obtained by using single-objective solution methods (such as genetic algorithm or particle swarm algorithm), and the results are shown in Table 2.

[0105] Table 2 Optimization schemes for maximizing single-objective returns

[0106]

[0107] The single-objective solution method was used to obtain the scheme corresponding to minimizing the single-objective seedling damage risk, and the results are shown in Table 3.

[0108] Table 3 Optimization schemes corresponding to minimizing the risk of seedling damage in a single objective

[0109]

[0110] The single-objective solution method was used to obtain the scheme corresponding to minimizing the single-objective lodging risk, and the results are shown in Table 4.

[0111] Table 4 Optimization schemes for minimizing single-target lodging risk

[0112]

[0113] Compared to single-objective profit maximization, the third solution of the Pareto method yields 1.5% less profit, but reduces the risk of landslide by 27%.

[0114] Compared to minimizing the risk of seedling damage in a single objective, the third solution of the Pareto method has a 6% increased risk of lodging, but yields 10% more profit.

[0115] Compared to minimizing the single-objective lodging risk, the first solution of the Pareto method has roughly the same lodging risk, but yields 297% more profit.

[0116] Therefore, the solution method of this invention has the ability to perform multi-objective collaborative optimization, while taking into account the benefits, seedling damage risk and lodging risk; through ecological coupling modeling, it integrates crab-rice interaction, environmental factors and nonlinear relationships; and it can be directly applied to production practice, outputting planting density, variety selection and breeding strategies.

[0117] See Figure 2This embodiment also provides a crab-rice ratio decision-making system based on multi-objective optimization, including: a data acquisition and storage module, a yield prediction module, a multi-objective optimization modeling module, an optimization solution module, and a feedback update module. The data acquisition and storage module is used to collect and store historical crab-rice co-cultivation production data and environmental parameters. The historical crab-rice co-cultivation production data includes crab seedling release density, rice planting density, rice variety, initial crab size, crab release batches, crab yield, and rice yield. The environmental parameters include ditch ratio, average water depth, temperature, and dissolved oxygen level. The yield prediction module is used to construct rice yield prediction models and crab yield prediction models based on the historical crab-rice co-cultivation production data and environmental parameters. The multi-objective optimization modeling module is used to construct a multi-objective optimization model that includes maximizing comprehensive economic benefits, minimizing rice seedling damage risk, and minimizing rice lodging risk. The optimization module is used to set constraints and iteratively solve the multi-objective optimization model, outputting: the optimized crab seedling release density, rice planting density, corresponding rice varieties, crab specifications and release batches, and the values ​​of each objective function under the optimal solution. The constraints include the range constraints of crab seedling release density and rice planting density, physical condition constraints of ditch ratio, average water depth and dissolved oxygen, and risk threshold constraints set based on the occurrence of lodging and seedling damage over the years. The feedback update module is used to update the historical crab-rice co-cultivation production data, rice and crab yield prediction model and multi-objective optimization model based on the actual production data of the optimal solution.

[0118] Finally, it should be noted that the above embodiments are intended to illustrate the technical solutions of the present invention and do not constitute any limitation on the present invention. Those skilled in the art should fully understand that modifications to the technical solutions described in the foregoing embodiments or equivalent substitutions for any part or all of the technical features are entirely feasible. Such modifications or substitutions, as long as they do not depart from the scope of protection defined by the claims of the present invention, should be considered reasonable extensions of the present invention.

Claims

1. A crab-rice ratio decision-making method based on multi-objective optimization, characterized in that, include: Step S1: Collect historical crab-rice co-cultivation production data and environmental parameters; the historical crab-rice co-cultivation production data includes crab seedling release density, rice planting density, rice variety, initial crab size, crab release batch, crab yield, and rice yield; the environmental parameters include ditch ratio, average water depth, temperature, and dissolved oxygen level; Step S2: Preprocess the historical crab-rice co-cultivation production data and environmental parameters collected in Step S1; Step S3: Based on the preprocessed historical rice-crab co-cultivation production data and environmental parameters, construct rice yield prediction models and crab yield prediction models respectively; Step S4: Based on the output results of the rice yield prediction model and the crab yield prediction model, construct a multi-objective optimization model for the crab-rice ratio that includes the objectives of maximizing comprehensive economic benefits, minimizing feed costs, minimizing rice seedling damage risk, and minimizing rice lodging risk. Step S5: Introduce constraints into the multi-objective optimization model and perform iterative solutions to obtain the optimal solution set that satisfies the constraints; Step S6: Select the optimal solution from the set of optimal solutions according to decision preferences and output it as the crab-rice ratio optimization scheme; the output results include: the optimized crab seedling release density, rice planting density, corresponding rice varieties, crab specifications and release batches, and the values ​​of each objective function under the optimal solution.

2. The crab-rice ratio decision-making method based on multi-objective optimization according to claim 1, characterized in that, In step S1, the historical rice-crab co-cultivation production data also includes average plant height, feed cost, risk of seedling damage, and risk of lodging; wherein, the risk of seedling damage is the proportion of seedling loss area caused by crab activity to the planted area; the risk of lodging is the proportion of rice lodging area to the planted area; and the crab seedling release density is defined as the number of crabs released per unit area of ​​rice field, in units of crabs / hm². 2 Rice planting density is defined as the number of rice plants per unit area of ​​paddy field, expressed as plants per hectare (hm²). 2 .

3. The crab-rice ratio decision-making method based on multi-objective optimization according to claim 2, characterized in that, In step S2, the preprocessing of the historical crab-rice co-cultivation production data and environmental parameters collected in step S1 specifically includes: The collected historical rice-crab co-cultivation production data underwent missing value processing, outlier removal, unit unification, standardization, and feature coding. Specifically, discrete variables such as rice variety, initial crab size, and release batch were encoded using one-hot or ordinal coding. Continuous variables, including ditch ratio, average water depth, temperature, dissolved oxygen, rice planting density, and crab release density, were normalized or standardized to construct a numerical feature matrix. Simultaneously, rice yield, crab yield, feed cost, lodging risk, and seedling damage risk were used as target variables to construct corresponding label vectors.

4. The crab-rice ratio decision-making method based on multi-objective optimization according to claim 2, characterized in that, In step S3, based on the pretreated rice yield Corresponding crab seedling stocking density Rice planting density Rice varieties Ditch ratio ,temperature Average water depth Dissolved oxygen Based on the relationship between the two factors, a rice yield prediction model is constructed as follows: in, ~ This represents the correlation coefficient.

5. The crab-rice ratio decision method based on multi-objective optimization according to claim 4, characterized in that, In step S3, based on the pre-processed crab yield With the density of crab seedlings Rice planting density Ditch ratio ,temperature Average water depth Dissolved oxygen Based on the relationship between the two factors, the following model for predicting crab production is constructed: in, ~ This represents the correlation coefficient.

6. The crab-rice ratio decision method based on multi-objective optimization according to claim 5, characterized in that, In step S4, the multi-objective optimization model includes the following objective function: 1) Objective function for maximizing overall economic benefits: in, For rice profits; For the benefit of river crabs; For feed costs; This is the market price of rice, in yuan / kg. Rice yield, unit: kg / hm 2 ; The price listed is the market price of river crabs, in yuan / kg. River crab yield, unit: kg / hm 2 ; This is the unit price of feed, in yuan / kg; To match the initial size of the river crab and batches of river crabs released The relevant average weight gain, ; To match the specifications of river crabs and batches of river crabs released The relevant feed conversion ratio, ; Survival rate; Obtained by fitting historical crab-rice co-cultivation production data; 2) Objective function for minimizing feed costs: 3) Objective function for minimizing the risk of rice seedling damage: in, Average water depth, unit: m; This represents the proportion of the ditch area; Dissolved oxygen, unit: mg / L; ~ These are the fitting coefficients obtained by fitting historical crab-rice co-cultivation production data; 4) Objective function for minimizing the risk of rice lodging: in, This represents the average plant height. ~ These are the fitting coefficients obtained by fitting historical data; The objectives of maximizing overall economic benefits, minimizing feed costs, minimizing the risk of rice seedling damage, and minimizing the risk of rice lodging, as mentioned above, together constitute the objective function vector: The decision variable vector is: .

7. The crab-rice ratio decision method based on multi-objective optimization according to claim 6, characterized in that, In step S5, the constraint conditions include at least one of the following conditions: Crab seedling stocking density With rice planting density The range of values ​​is constrained; Ditch ratio Average water depth With dissolved oxygen Physical constraints; Risk threshold constraints were set based on the occurrence of lodging and seedling damage over the years.

8. The crab-rice ratio decision method based on multi-objective optimization according to claim 1, characterized in that, In step S5, the solution methods include: combinatorial optimization algorithms, multi-objective evolutionary algorithms, mathematical programming methods, heuristic search algorithms, or any combination thereof.

9. The crab-rice ratio decision-making method based on multi-objective optimization according to claim 1, characterized in that, The method also includes: after the implementation of the crab-rice ratio optimization scheme obtained in S6, collecting production data and result data in the actual production process and entering them into the historical crab-rice co-cultivation production database; and updating the rice yield prediction model, crab yield prediction model and multi-objective optimization model based on the updated historical crab-rice co-cultivation production data.

10. A crab-rice ratio decision system based on multi-objective optimization, used to implement the crab-rice ratio decision method based on multi-objective optimization as described in any one of claims 1 to 9, characterized in that, include: The data acquisition and storage module is used to collect and store historical crab-rice co-cultivation production data and environmental parameters; The yield prediction module is used to construct rice yield prediction models and crab yield prediction models based on the historical rice-crab co-cultivation production data and environmental parameters. The multi-objective optimization modeling module is used to construct a multi-objective optimization model that includes maximizing comprehensive economic benefits, minimizing the risk of rice seedling damage, and minimizing the risk of rice lodging. The optimization solution module is used to iteratively solve the multi-objective optimization model and output the optimal solution set that satisfies the constraints and the corresponding objective function values ​​under the solution. The feedback update module is used to update historical rice-crab co-cultivation production data, rice and crab yield prediction models, and multi-objective optimization models based on the actual production data of the optimal solution.

Images