An intelligent crop planting decision method and system

Abstract

The present disclosure provides a kind of crop intelligent planting decision method and system, it is related to agricultural planting planning and intelligent decision-making technical field, including: obtaining the plot information of target area, optional crop information and historical planting sales record;With the maximization of total profit of target area as the goal, construct decision model, and increase the constraint of plot arable type, crop planting rule to planting condition;Price fluctuation and yield fluctuation are modeled, and price yield random sequence is generated;Price yield random sequence is used as risk scenario set, and the decision model under different scenarios is solved by nested genetic algorithm, to maximize the total profit in the worst case as the goal, select the final risk-resistant planting strategy from the obtained several planting strategies.The present application is suitable for crop planting planning in arable land resource limited areas (such as mountainous areas, dry land), supports the maximization of income and risk control under multiple constraints.

Description

Technical Field

[0001] This disclosure relates to the field of agricultural planting planning and intelligent decision-making technology, specifically to an intelligent crop planting decision-making method and system. Background Technology

[0002] In the context of sustainable rural economic development, making rational use of limited arable land resources and adopting appropriate crop planting strategies are crucial for improving agricultural production efficiency and reducing planting risks. This is especially true for areas with limited natural conditions, where the importance of such strategies is self-evident.

[0003] Existing crop planting decision-making methods have the following shortcomings, which prevent the acquisition of optimal planting strategies and affect the final planting results:

[0004] (1) Static programming: Traditional models rely on fixed parameters (such as stable prices and output), which cannot cope with market fluctuations and climate risks.

[0005] (2) Single objective optimization: Focusing only on profit maximization and ignoring long-term factors such as crop rotation constraints (e.g., the planting cycle of legumes) and soil health.

[0006] (3) High computational complexity: Large-scale planting problems involve thousands of variables, which are difficult to solve efficiently using traditional linear programming.

[0007] (4) Lack of risk resistance: The impact of price and output uncertainty on revenue is not considered. Summary of the Invention

[0008] To address the aforementioned issues, this disclosure proposes an intelligent crop planting decision-making method and system, applicable to crop planting planning in areas with limited arable land resources (such as mountainous and arid land), supporting profit maximization and risk control under multiple cycles and constraints.

[0009] According to some embodiments, the present disclosure adopts the following technical solutions:

[0010] A smart crop planting decision-making method includes:

[0011] Obtain information on land parcels, available crops, and historical planting and sales records for the target area;

[0012] Based on land parcel information and crop information, with the goal of maximizing the total profit of the target area, a decision model is constructed with the planting of the j-th crop on the i-th land parcel in the k-th year as the decision variable, and the constraints of land parcel type and crop planting pattern on planting conditions are added.

[0013] Based on historical planting and sales records, uncertainty modeling is performed on price and yield fluctuations to generate random price and yield sequences;

[0014] Using a random sequence of price and output as a set of risk scenarios, a nested genetic algorithm is used to solve the decision-making model under different scenarios, optimize the planting strategy under each scenario, and select the final risk-resistant planting strategy from several obtained planting strategies with the goal of maximizing the total profit under the worst case.

[0015] According to some embodiments, the present disclosure adopts the following technical solutions:

[0016] A smart crop planting decision-making system includes:

[0017] The information acquisition module is configured to acquire land parcel information, selectable crop information, and historical planting and sales records for the target area.

[0018] The model building module is configured to: based on land parcel information and crop information, with the goal of maximizing the total profit of the target area, construct a decision model with the planting of the jth crop on the i-th land parcel in the k-th year as the decision variable, and add constraints on the planting conditions by the land parcel type and crop planting patterns;

[0019] The sequence generation module is configured to: perform uncertainty modeling on price and yield fluctuations based on historical planting and sales records, and generate random price and yield sequences;

[0020] The nested optimization module is configured to: use the random sequence of price and output as a set of risk scenarios, solve the decision model under different scenarios through a nested genetic algorithm, optimize the planting strategy under each scenario, and select the final risk-resistant planting strategy from several obtained planting strategies with the goal of maximizing the total profit under the worst case.

[0021] According to some embodiments, the present disclosure adopts the following technical solutions:

[0022] A computer program product includes a computer program that, when executed by a processor, implements the aforementioned intelligent crop planting decision-making method.

[0023] According to some embodiments, the present disclosure adopts the following technical solutions:

[0024] A non-transitory computer-readable storage medium is provided for storing computer instructions, which, when executed by a processor, implement the aforementioned intelligent crop planting decision-making method.

[0025] According to some embodiments, the present disclosure adopts the following technical solutions:

[0026] An electronic device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the aforementioned intelligent crop planting decision-making method.

[0027] Compared with the prior art, the beneficial effects of this disclosure are as follows:

[0028] This invention addresses changes in extreme conditions by constructing a set of risk scenarios and decision-making models for individual scenarios. With the goal of maximizing total profit under the worst-case scenario, it seeks the optimal planting strategy with the strongest risk resistance and the best benefits. This facilitates field management, improves production efficiency, and reduces planting risks that may be caused by various uncertainties. Attached Figure Description

[0029] The accompanying drawings, which form part of this disclosure, are used to provide a further understanding of this disclosure. The illustrative embodiments of this disclosure and their descriptions are used to explain this disclosure and do not constitute an undue limitation of this disclosure.

[0030] Figure 1 This is a flowchart of an intelligent crop planting decision-making method as shown in Example 1. Detailed Implementation

[0031] The present disclosure will be further described below with reference to the accompanying drawings and embodiments.

[0032] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of this disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains.

[0033] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this disclosure. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms “comprising” and / or “including” are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0034] Example 1

[0035] One embodiment of this disclosure provides an intelligent crop planting decision-making method. Taking a village in the mountainous region of North China as an example, it plans crop planting strategies for the available arable land in the village, such as... Figure 1 As shown, the specific steps are as follows:

[0036] Step S1: Obtain land parcel information, available crop information, and historical planting and sales records for the target area.

[0037] The farmland in the target area is divided into plots of varying sizes. Each plot has attributes of area and farmland type. The farmland types include open farmland (flat dry land, terraced fields, hillside land, irrigated land) and facility farmland (ordinary greenhouses, smart greenhouses).

[0038] Crops are classified according to their planting patterns as follows:

[0039] o Grains: Legumes (soybeans, black beans, etc.), staple grains (wheat, corn, etc.), rice.

[0040] o Vegetables: leafy greens (spinach, romaine lettuce), root vegetables (radish), legumes (cowpeas).

[0041] Edible fungi: elm mushrooms, shiitake mushrooms, etc.

[0042] Among them, flat dry land, terraced fields and hillsides are suitable for planting one season of grain crops per year, while irrigated land is suitable for planting one season of rice or two seasons of vegetables per year; ordinary greenhouses are suitable for planting one season of vegetables and one season of edible fungi per year, while smart greenhouses are suitable for planting two seasons of vegetables per year.

[0043] Historical planting and sales records mainly include the yield per mu, sales price, planting cost, and sales volume of the j-th crop planted on plot i in year k.

[0044] Specifically, the village has unique geographical and climatic conditions, with consistently low temperatures, resulting in most of its arable land only being suitable for one crop per year. Specifically, the village has 1201 mu (approximately 87 hectares) of open farmland, divided into 34 plots of varying sizes, including flat land, terraced fields, hillside land, and irrigated land. Flat land, terraced fields, and hillside land are suitable for growing one crop of grain crops; while irrigated land is suitable for growing one crop of rice or two crops of vegetables. In addition, the village possesses agricultural facilities, including 16 ordinary greenhouses and 4 smart greenhouses, each covering 0.6 mu (approximately 0.04 hectares). Ordinary greenhouses are suitable for growing one crop of vegetables and one crop of edible fungi per year, while the more advanced smart greenhouses are suitable for growing two crops of vegetables per year. Furthermore, according to the growth patterns of crops, each crop cannot be continuously planted on the same plot, otherwise it will lead to reduced yields. Legumes should also be planted at least once every three years to improve the soil. Choosing suitable crops based on local conditions and optimizing planting strategies can not only facilitate field management but also effectively improve production efficiency.

[0045] Obtain three databases related to crop cultivation in the village: arable land database (plot type, area, historical planting records), market database (real-time price, historical volatility, demand forecast), and crop parameter database (yield per acre, cost, crop rotation rules). Based on the known historical planting and sales records of the village from 2020 to 2023, plan a seven-year planting strategy from 2024 to 2030.

[0046] Step S2: Based on land parcel information and crop information, with the goal of maximizing the total profit of the target region, construct a decision model with the planting of the j-th crop on the i-th land parcel in year k as the decision variable, and add constraints on planting conditions based on land type and crop planting patterns. The specific steps are as follows:

[0047] 1. Variable definition and constraint modeling

[0048] Decision variable: Define a 0-1 variable x kij This represents a set of 0-1 variables x representing whether the i-th plot of land in year k is planted with the j-th crop, and both plots belong to year k. kij The planting strategy X for year k is formed. k .

[0049] Core constraints:

[0050] (1) Each plot of land can only be planted with one type of crop per year, as expressed by the formula:

[0051]

[0052] (2) Crops can only be planted on plots of land that are suitable for the type of arable land, such as rice which can only be planted on irrigated land.

[0053] (3) The same crop cannot be planted on the same plot of land for two consecutive years, as expressed by the formula:

[0054] x kij +x (k+1)ij ≤1

[0055] (4) Legumes should be planted at least once every three years, as expressed by the formula:

[0056]

[0057] 2. Construction of the objective function

[0058] (1) Profit per plot:

[0059]

[0060] (2) Losses from unsold inventory: Excess quantities are recorded as losses through destruction or discounting, expressed by the following formula:

[0061]

[0062] Where λ = 1 (destruction) or 0.5 (discount).

[0063] (3) Formula for total profit:

[0064]

[0065] Among them, X k Let be the planting strategy for year k.

[0066] Specifically, for this village, the expected sales volume, planting cost, yield per acre, and selling price of various crops will remain stable compared to 2023, with crops planted each season to be sold in the same season. Two options are provided for yields exceeding expected sales: ① The excess will be completely destroyed and wasted. ② The excess will be sold at 50% of the 2023 selling price.

[0067] Step S3: Based on historical planting and sales records, perform uncertainty modeling on price and yield fluctuations to generate random price and yield sequences.

[0068] The expected sales volume, yield, planting cost, and sales price of crops are all uncertain, which may lead to market fluctuations and planting risks based on crops. Therefore, fluctuations are added to these factors in the crop planting process according to actual conditions. This embodiment takes yield and (sales) price as examples to generate a set of risk scenarios.

[0069] Specifically, a price fluctuation curve is fitted based on historical planting and sales records. and production fluctuation curve For example, the standard deviation of vegetable prices σ = 15%μ, and drought leads to a 20%-30% decrease in yield per acre.

[0070] Based on the two curves, Monte Carlo simulation was used to generate 1000 sets of price-output random sequences. Each generated price-output combination was treated as a scenario, forming a risk scenario set.

[0071] Step S4: Using the price-yield random sequence as a risk scenario set, a nested genetic algorithm is used to solve the decision-making model under different scenarios, optimizing the planting strategy under each scenario, with the goal of maximizing the total profit in the worst case. The final risk-resistant planting strategy is selected from the obtained planting strategies. First, the risk scenario set is explained:

[0072] Define a set of discrete risk scenarios Each scenario contains an extreme combination of the following random parameters:

[0073] Production fluctuation: ±ΔY j (e.g., a 20% reduction in production due to a natural disaster)

[0074] Price fluctuation: ±ΔP j (For example, market saturation leading to a 30% price drop)

[0075] Cost increase: +ΔC j (For example, fertilizer prices increase by 15%)

[0076] Changes in demand: ±ΔQ j (If the emergence of alternatives leads to a decline in expected sales)

[0077] Unsold inventory loss rule: Sell at 50% of the original price or destroy completely (scenario-dependent).

[0078] Example scenarios are shown in Table 1:

[0079] Table 1 Risk Scenarios

[0080]

[0081] 1. Traverse each scenario in the risk scenario set, that is, under the price-output combination, with the goal of maximizing the total profit of the target region, solve the decision model constructed in step S2 to obtain the optimal planting strategy under each scenario.

[0082] This embodiment uses a genetic algorithm to solve the decision model, specifically:

[0083] First, encoding is performed, concatenating 0-1 variables into a chromosome, such as 1201 mu of arable land × 7 years × 15 crops = 126,105-dimensional vector.

[0084] Then, the fitness function is defined as the negative of the objective function value (total profit) (since the genetic algorithm minimizes it by default), expressed by the formula:

[0085]

[0086] Among them, X kij ∈{0,1} is the decision variable, representing whether crop j is planted on plot i in year k; Y kij The yield per mu (catties / mu) of the crop planted on plot i in year k; P kj Let A be the selling price per kilogram of crop j in year k; i C represents the area (in mu) of plot i; kj L represents the unit area planting cost (yuan / mu) of crop j in year k; kij The losses were due to slow sales.

[0087] Secondly, set the parameters: population size = 500, crossover probability = 0.8 (two-point crossover), mutation probability = 0.01 (single-point mutation), and termination condition = 200 iterations or fitness change rate < 1‰.

[0088] Finally, based on the above coding, definition, and settings, the planting strategy is iteratively solved until the termination condition is met, and the optimal planting strategy for each scenario is obtained.

[0089] Here is a specific example of using a genetic algorithm to solve for the optimal planting strategy, including the following steps:

[0090] Step 1: Initialization and Data Input

[0091] 1) Construction of arable land database

[0092] Enter the type of land (flat land, terraced fields, etc.), area, historical planting records, and define the attributes of crops: yield per acre, cost price, sales price range, and strategies for handling unsold crops.

[0093] 2) Constraint rule base generation, including:

[0094] Strict constraints: single-crop planting in the same season on the same plot of land; continuous planting of the same crop is prohibited.

[0095] Soft constraints: the number of times beans are planted within three years must be ≥1, and the number of seasons for irrigated land is limited.

[0096] Step 2: Chromosome hierarchical coding

[0097] 1) Three-level coding structure design, including:

[0098] The plot layer assigns an independent gene segment to each plot.

[0099] The time layer divides gene segments by year to mark planting history;

[0100] The crop layer uses one-hot coding to represent crop selection (e.g., 0010 represents the third crop).

[0101] 2) Gene compression optimization

[0102] For restricted land parcels such as hillsides, illegal crop coding bits (e.g., rice coding) are masked, and an initial population size of N=200 is generated, satisfying the following conditions: (D = 6,188 dimensions).

[0103] Step 3: Dynamic assessment of fitness

[0104] 1) Multi-objective return calculation, including:

[0105] Basic profit: P = ∑(sales revenue - planting costs - penalties for unsold goods);

[0106] Penalties for unsold goods: Excess quantity will be destroyed (100% loss) or discounted (50% loss);

[0107] Soil health bonus: Each completed legume rotation increases the yield of subsequent crops by 2%.

[0108] 2) Implement tiered penalties for violations, including:

[0109] Violation of hard constraints (such as duplicate planting): Directly eliminate the individual;

[0110] Soft constraint violation (e.g., bean spacing): Apply penalty value = 10 4 ×(3 - actual number of plantings).

[0111] Step 4: Evolutionary Operation

[0112] 1) Tournament Selection

[0113] 5% of individuals are randomly selected to form a competition group, and the individuals with the best fitness are retained to enter the next generation.

[0114] 2) Targeted crossover strategy

[0115] Gene exchange is preferentially performed at safe sites for crop rotation (such as non-prohibited sites after legume planting), with adaptive adjustment of crossover probability: p c =0.8×e -0.01t (t is the number of iterations).

[0116] 3) Constraint-perceived variation

[0117] Apply high mutation probability (p) to consecutive planting sites m =0.1), and a protection mechanism (p) is adopted for compliant gene segments. m =0.001).

[0118] Step 5: Parallel Acceleration and Convergence Determination

[0119] 1) GPU parallel computing

[0120] The population assessment task was distributed across 8 computational cores, and fitness values ​​were synchronized in real time.

[0121] 2) Dual termination conditions

[0122] Absolute convergence: The improvement of the optimal solution is less than 0.1% for 50 consecutive iterations; Forced termination: Total number of iterations > 500.

[0123] 2. With the goal of maximizing total profit under the worst-case scenario, select the planting strategy with the strongest risk resistance and the best benefits from the several planting strategies obtained.

[0124] The worst-case scenario here represents the planting strategy that minimizes total profit across all scenarios. As long as the total profit is maximized in the worst-case scenario, the risk can be mitigated. Therefore, the objective function is defined as:

[0125]

[0126] Unsold inventory loss (X) k ,ξ)=∑ j max(0,∑ i Y ij (ξ)·X kij -Q j (ξ))·α(ξ)

[0127] Among them, X k Let Y be the planting strategy for year k, ξ be the scenario with concentrated risk, and Y be the planting strategy for year k. ij (ξ) represents the yield of crop j planted on plot i in scenario ξ, P j (ξ) represents the selling price of crop j in scenario ξ, and C ij (ξ) represents the cost of planting crop j on plot i under scenario ξ, and α(ξ) represents the unsold loss coefficient under scenario ξ (e.g., α = 1 when there is no unsold stock, and α = 0.5 when sold at a discount).

[0128] Dynamic rotation constraints and resource limitations are added during the solution process, specifically:

[0129] (1) Dynamic rotation constraint

[0130] Introduce scenario-dependent crop rotation rules (such as fallowing after extreme weather):

[0131] X kij +X (k+1)ij ≤1 if ξ includes soil degradation risk

[0132] (2) Resource constraints

[0133] Adjusting the upper limit of planting area due to cost fluctuations:

[0134]

[0135] Under the aforementioned objective function and constraints, a two-level optimization framework is used to select the planting strategy with the strongest risk resistance and the best benefits from several obtained planting strategies, specifically as follows:

[0136] Inner layer optimization (for fixed planting strategy X) k ): Traverse all scenes Calculate the profit in the worst-case scenario, i.e. Specifically, a genetic algorithm is used to nest scenario traversal within the fitness function to calculate the worst-case profit for each individual (planting strategy).

[0137] Outer layer optimization: Adjust planting strategy X k Maximize the worst-case profit obtained from the inner layer, i.e. Specifically, a column generation method is adopted to decompose the problem into a "main problem (planting strategy)" and a "sub-problem (scenario evaluation)" for large-scale scenario sets.

[0138] Example 2

[0139] One embodiment of this disclosure provides an intelligent crop planting decision-making system, including:

[0140] The information acquisition module is configured to acquire land parcel information, selectable crop information, and historical planting and sales records for the target area.

[0141] The model building module is configured to: based on land parcel information and crop information, with the goal of maximizing the total profit of the target area, construct a decision model with the planting of the jth crop on the i-th land parcel in the k-th year as the decision variable, and add constraints on the planting conditions by the land parcel type and crop planting patterns;

[0142] The sequence generation module is configured to: perform uncertainty modeling on price and yield fluctuations based on historical planting and sales records, and generate random price and yield sequences;

[0143] The nested optimization module is configured to: use the random sequence of price and output as a set of risk scenarios, solve the decision model under different scenarios through a nested genetic algorithm, optimize the planting strategy under each scenario, and select the final risk-resistant planting strategy from several obtained planting strategies with the goal of maximizing the total profit under the worst case.

[0144] Example 3

[0145] One embodiment of this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the aforementioned intelligent crop planting decision-making method.

[0146] Example 4

[0147] One embodiment of this disclosure provides a non-transitory computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the aforementioned intelligent crop planting decision-making method.

[0148] Example 5

[0149] One embodiment of this disclosure provides an electronic device, including a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the aforementioned intelligent crop planting decision-making method.

[0150] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0151] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0152] While the specific embodiments of this disclosure have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of this disclosure. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of this disclosure are still within the scope of protection of this disclosure.

Claims

1. A smart crop planting decision-making method, characterized in that, The method comprises the following steps: acquiring plot information, optional crop information, and historical planting and sales records of a target region; Based on the plot information and crop information, a decision model is constructed with the total profit maximization of the target area as the target, taking the planting of the first crop in the plot as the decision variable, and adding the constraints of the plot type, crop planting rules and planting conditions. In the year The first crop is planted in the plot The decision model is constructed, and the constraints of the plot type, crop planting rules and planting conditions are added. performing uncertainty modeling on price fluctuations and yield fluctuations based on the historical planting and sales records to generate a price-yield random sequence; using the price-yield random sequence as a risk scenario set, solving a decision model under different scenarios by a nested genetic algorithm, optimizing planting strategies under each scenario, and selecting a final risk-resistant planting strategy from the obtained planting strategies, with the goal of maximizing the total profit in the worst case; wherein the nested genetic algorithm specifically comprises: iterating through each scenario in the risk scenario set, solving the constructed decision model under the price-yield combination with the goal of maximizing the total profit of the target region, and obtaining the optimal planting strategy under each scenario; selecting a planting strategy with the strongest risk resistance and the best benefit from the obtained planting strategies, with the goal of maximizing the total profit in the worst case; the worst case is the planting strategy with the smallest total profit in all scenarios, as long as the total profit in the worst case is maximized, the risk can be resisted, therefore, the objective function is defined as: wherein, is the planting strategy for the year, is a scenario from a set of risk scenarios, is the scenario is the field is the yield of the crop planted, is the selling price of the crop is the cost of planting the crop in the field is the expected sales volume of the crop is the decision variable representing whether the field is the loss coefficient for unsold inventory in scenario ξ;​​​​​​​​​​​ adding dynamic crop rotation constraints and resource limitation constraints in the solving process, specifically comprising: (1) dynamic crop rotation constraints introducing scenario-dependent crop rotation rules: (2) resource limitation constraints adjusting the upper limit of planting area under cost fluctuations: under the above objective function and constraints, a double-layer optimization framework is used to select a planting strategy with the strongest risk resistance and the best benefit from the obtained planting strategies, specifically comprising: Inner loop optimization: for fixed planting strategy , iterate over all scenarios , compute the worst-case profit, i.e. , in particular, employ a genetic algorithm, with scenario iteration nested in the fitness function, to compute the worst-case profit for each individual; Outer optimization: adjust planting strategy , maximize the worst profit obtained by inner optimization, i.e. , specifically, adopt column generation method, decompose the problem into master problem planting strategy and sub-problem scenario evaluation for large-scale scenario set.

2. The method of claim 1, wherein the crop intelligent planting decision is determined based on the crop growth model and the crop growth environment data. the constraints of the plot cultivation type and crop planting rules on planting conditions include: each plot can only plant one crop per year; crops can only be planted on plots with suitable cultivation types; the same plot cannot plant the same crop for two consecutive years; beans must be planted at least once every three years.

3. The intelligent crop planting decision-making method as described in claim 1, characterized in that, The uncertainty modeling of price fluctuations and yield fluctuations is based on historical planting and sales records to fit price fluctuation curves and yield fluctuation curves.

4. The method of claim 3, wherein the crop intelligent planting decision is determined by the steps of: determining a crop planting area; determining a crop planting density; determining a crop planting pattern; and determining a crop planting direction. The price-yield random sequence is generated based on the price fluctuation curve and the yield fluctuation curve through Monte Carlo simulation.

5. A system for implementing the method of intelligent crop plantation decision as claimed in any one of claims 1 to 4, wherein, The method comprises the following steps: an information acquisition module configured to acquire plot information, optional crop information, and historical planting and sales records of a target region; The model construction module is configured to: based on the plot information and the crop information, construct a decision model with the plot planting the crop as a decision variable, and increase the constraints of the plot cultivation type and the crop planting rule on the planting condition, with the target of maximizing the total profit of the target region. the year The plot plants the crop as a decision variable, and increases the constraints of the plot cultivation type and the crop planting rule on the planting condition. ​ a sequence generation module configured to perform uncertainty modeling on price fluctuations and yield fluctuations based on the historical planting and sales records to generate a price-yield random sequence; a nested optimization module configured to use the price-yield random sequence as a risk scenario set, solve a decision model under different scenarios by a nested genetic algorithm, optimize planting strategies under each scenario, and select a final risk-resistant planting strategy from the obtained planting strategies, with the goal of maximizing the total profit in the worst case.

6. A computer program product comprising a computer program, characterized in that, The computer program is executed by a processor to implement the crop intelligent planting decision method of any one of claims 1-4.

7. A non-transitory computer-readable storage medium, comprising: The non-transitory computer readable storage medium is used to store computer instructions, which are executed by a processor to implement the crop intelligent planting decision method of any one of claims 1-4.

8. An electronic device, comprising: The method comprises the following steps: A processor, a memory and a computer program; wherein the processor is connected with the memory, the computer program is stored in the memory, when the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes the method for realizing the intelligent planting decision of the crop according to any one of claims 1-4.

Images