Agricultural resource allocation method and system based on global optimization

Abstract

The present application relates to the field of agricultural information technology and intelligent decision-making technology, and relates to a kind of agricultural resource allocation method and system based on global optimization, comprising: the target farmland area is divided into a farmland subarea, data acquisition module is controlled by agricultural resource allocation system, and a plurality of agricultural data sets are obtained by data acquisition to a farmland subarea, a plurality of effective agricultural data sets are obtained by preprocessing, and a fractional structure optimization model is constructed by using optimization model construction module, a weighted target function is obtained by weighting processing, a single-objective solution model is obtained by combining the weighted target function and the joint constraint condition system, and the single-objective solution model is optimized and solved by global optimization solving module, to obtain a globally optimal configuration scheme.The present application can eliminate the resource mismatch problem in agricultural resource allocation, adapt to the multi-dimensional resource constraint scene of multi-subarea contiguous large-scale farmland, and maximize the efficiency of agricultural resource utilization.

Description

Technical Field

[0001] This invention relates to the fields of agricultural information technology and intelligent decision-making technology, and in particular to an agricultural resource allocation method and system based on global optimization. Background Technology

[0002] With the rapid development of the Internet of Things, big data, and smart agriculture, the global optimization and allocation of agricultural resources has become a core technology for modern agricultural production management and food security, directly affecting the efficiency of agricultural resource utilization. Currently, the coordinated allocation of resources across multiple zones in large-scale, contiguous farmland faces severe challenges such as water scarcity and resource misallocation. Therefore, constructing a globally optimal, multi-constraint adaptive, and scientifically quantifiable agricultural resource optimization and allocation system has become an urgent problem to be solved.

[0003] Traditional agricultural resource allocation often relies on subjective experience and historical planting habits, distributing various resources equally according to farmland area. Some schemes adopt simple single-objective linear programming, considering only a single resource constraint or output indicator.

[0004] While the aforementioned methods can achieve basic resource allocation, they lack quantitative analysis of the differences in production capacity and resource consumption among different farmland zones. This easily leads to resource shortages in high-yield, low-consumption areas and resource waste in low-yield, high-consumption areas, resulting in low overall utilization efficiency. Furthermore, they fail to simultaneously achieve the dual objectives of maximizing output and minimizing consumption, and are prone to getting stuck in local optima, making it difficult to obtain a truly globally optimal solution. Therefore, this invention proposes an agricultural resource allocation method and system based on global optimization to address the aforementioned shortcomings of existing technologies. Summary of the Invention

[0005] This invention provides an agricultural resource allocation method based on global optimization. Its main purpose is to eliminate the resource mismatch problem in agricultural resource allocation, adapt to the multi-dimensional resource constraint scenarios of large-scale farmland with multiple zones, and ensure the maximization of agricultural resource utilization efficiency.

[0006] To achieve the above objectives, the present invention provides an agricultural resource allocation method based on global optimization, comprising:

[0007] The system receives agricultural resource allocation instructions and determines an agricultural resource allocation system and target farmland area based on the instructions. The agricultural resource allocation system includes a data acquisition module, an optimization model construction module, and a global optimization solution module.

[0008] The target farmland area is divided into a farmland partitions. Data is collected from the a farmland partitions through the data acquisition module to obtain a multi-source agricultural datasets. Each multi-source agricultural dataset contains m multi-source agricultural data points labeled with the collection time. Each multi-source agricultural data point includes: crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planting area and land usable area.

[0009] Preprocessing a multi-source agricultural datasets yields a valid agricultural datasets, where the valid agricultural datasets, multi-source agricultural datasets, and farmland partitions correspond one-to-one.

[0010] By optimizing the model construction module, a fractional structure optimization model is constructed based on the a effective agricultural datasets, wherein the fractional structure optimization model includes a basic objective function and a joint constraint system;

[0011] Set output weight coefficients and consumption weight coefficients, and apply weighting to the basic objective function using these coefficients to obtain a weighted objective function. Combine the weighted objective function with the joint constraint system to obtain a single-objective solution model.

[0012] The single-objective solution model is optimized and solved by the global optimization solution module to obtain the globally optimal configuration scheme, thus completing the agricultural resource allocation based on global optimization.

[0013] Optionally, the preprocessing of a multi-source agricultural datasets to obtain a valid agricultural datasets includes:

[0014] Extract multi-source agricultural datasets sequentially from 'a' multi-source agricultural datasets, and perform the following operations on each multi-source agricultural dataset:

[0015] The multi-source agricultural dataset is converted into m feature vectors. The m feature vectors are then normalized to obtain m normalized feature vectors. Each normalized feature vector corresponds one-to-one with the multi-source agricultural data, and each feature vector includes k dimensional indicators.

[0016] For each of the m normalized feature vectors, perform the following operation:

[0017] Calculate the Euclidean distance between the normalized feature vector and the preset baseline feature vector to obtain the vector distance;

[0018] If the vector distance is not greater than the preset distance threshold, the multi-source agricultural data corresponding to the normalized feature vector is confirmed as valid agricultural data, and the valid agricultural data is summarized to obtain a valid agricultural dataset.

[0019] By summing up the valid agricultural datasets, we obtain a valid agricultural datasets.

[0020] Optionally, the step of constructing a fractional structure optimization model based on the a valid agricultural datasets by optimizing the model construction module includes:

[0021] Extract the valid agricultural datasets sequentially from the 'a' valid agricultural datasets, and perform the following operation on each valid agricultural dataset:

[0022] Extract n valid agricultural data points sequentially from the valid agricultural dataset;

[0023] Based on n valid agricultural data, calculate the average crop yield per unit area, average output conversion factor, average irrigation water consumption, average resource consumption factor, and average available land area of ​​the corresponding farmland zone, and then correlate them to obtain farmland zone nodes;

[0024] Summarize the farmland partition nodes to obtain a farmland partition nodes;

[0025] Set the planned planting area for each of the a farmland zones to obtain a planned planting area. Based on the planned planting area and the a farmland zone nodes, construct a joint constraint system through the optimization model construction module.

[0026] A basic objective function is pre-constructed, and a fractional structure optimization model is formed by the joint constraint system and the basic objective function.

[0027] Optionally, the step of calculating the average crop yield per unit area, average output conversion factor, average irrigation water consumption, average resource consumption factor, and average available land area of ​​the corresponding farmland zone based on n valid agricultural data includes:

[0028] From n valid agricultural data, we extract and summarize crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planting area and land usable area to obtain n crop yields, n crop output conversion factors, n farmland irrigation water consumption, n production resource consumption factor, n actual planting area and n land usable area.

[0029] Calculate the average output conversion factor for each of the n crops and the average usable land area for each of the n crops to obtain the average output conversion factor and the average usable land area.

[0030] The crop yields are extracted sequentially from n crop yields. Using the crop yields, the corresponding actual planting areas are retrieved from n actual planting areas. The ratio of crop yield to actual planting area is calculated to obtain the crop yield per unit area. The crop yields per unit area are then summarized to obtain n crop yields per unit area.

[0031] Calculate the average crop yield per unit area of ​​n units to obtain the average crop yield per unit area;

[0032] The average irrigation water consumption and average resource consumption coefficient are calculated based on n farmland irrigation water consumptions, n production resource consumption coefficients, and n actual planting areas.

[0033] Optionally, the step of constructing a joint constraint system based on a planned planting area and a farmland zoning nodes through an optimization model construction module includes:

[0034] Based on a planned planting area and a farmland partition nodes, single partition constraints are constructed by optimizing the model construction module;

[0035] From each of the a farmland partition nodes, extract and summarize the average irrigation water consumption, average resource consumption coefficient, and average available land area to obtain a average irrigation water consumption, a average resource consumption coefficient, and a average available land area.

[0036] Calculate the sum of a planned planting areas and the sum of a average available land areas to obtain the total planned planting area and total available land area of ​​the target farmland area.

[0037] Calculate the total irrigation water consumption based on a average irrigation water consumption and a planned planting area;

[0038] Calculate the total resource consumption based on a average resource consumption coefficient and a planned planting area;

[0039] Based on the total planned planting area, total available land area, total irrigation water consumption, and total resource consumption, a global total constraint is constructed through an optimization model construction module. The global total constraint is as follows:

[0040] ;

[0041] in, This represents the total planned planting area. This represents the total available land area. This represents the total irrigation water consumption. This indicates the preset irrigation water threshold. This represents the total resource consumption. This indicates the preset resource consumption threshold. Indicates the total quantity identifier. Indicates the default value identifier;

[0042] By summarizing the single-partition constraints and the global total constraints, a joint constraint system is obtained.

[0043] Optionally, the step of optimizing the single-objective solution model using a global optimization solution module to obtain a globally optimal configuration scheme includes:

[0044] Applying the Charnes-Cooper transform to the weighted objective function yields a linear objective function;

[0045] The joint constraint system is synchronously converted into a linear constraint system, and a standard linear programming model is obtained based on the linear objective function and the linear constraint system.

[0046] The feasible region of a linear programming problem is determined according to a standard linear programming model, wherein the feasible region of a linear programming problem includes multiple vertices;

[0047] Calculate the maximum objective function value corresponding to the feasible region of the linear programming problem, and calculate the optimal planting area to be planned based on the maximum objective function value;

[0048] Substitute the optimal planned planting area into the basic objective function to calculate the optimal unit resource output rate of the target farmland area.

[0049] By summarizing the optimal planned planting area and the optimal unit resource output rate, the globally optimal allocation scheme is obtained.

[0050] Optionally, the step of performing a Charnes-Cooper transform on the weighted objective function to obtain a linear objective function includes:

[0051] A global transformation variable and *a* transformed decision variables are introduced, wherein each transformed decision variable corresponds one-to-one with a farmland zone. The global transformation variable and *a* transformed decision variables are shown below:

[0052] ;

[0053] in, Represents the global transformation variable. This represents the consumption weighting coefficient. Indicates the first The average resource consumption coefficient of each farmland zone This represents the preset water resource consumption factor. Indicates the first Average irrigation water consumption per farmland zone Indicates the first The planned planting area for each farmland zone. Indicates the first Decision variables after transformation of each farmland zoning Indicates shared ownership One farmland zone;

[0054] Substituting the global transformation variable and the a transformed decision variables into the weighted objective function yields a linear objective function.

[0055] Optionally, the step of synchronously converting the joint constraint system into a linear constraint system includes:

[0056] Extract single-partition constraints and global total constraints from the joint constraint system;

[0057] Based on a transformed decision variables, global transformation variables, and a average available land area, the single-zone constraint is transformed into a linear single-zone constraint.

[0058] The linear total planting area, linear total irrigation water consumption, and linear total resource consumption are calculated based on a transformed decision variables, global transformation variables, a average irrigation water consumption, and a average resource consumption coefficient.

[0059] The global total constraint is converted into a linear total constraint based on the linear total planting area, linear total irrigation water consumption, and linear total resource consumption.

[0060] Linear normalization constraints are constructed based on the linear total resource consumption and consumption weight coefficients;

[0061] By summarizing linear single-partition constraints, linear total constraints, and linear normalization constraints, a system of linear constraint conditions is obtained.

[0062] Optionally, calculating the maximum objective function value corresponding to the feasible region of the linear programming problem includes:

[0063] Based on the linear constraint system, the current vertex is identified from multiple vertices;

[0064] The linear objective function value corresponding to the current vertex is calculated to obtain the current objective function value;

[0065] Based on the current vertex, identify a neighboring vertices from the feasible region of the linear programming. Calculate the linear objective function value of each of the a neighboring vertices to obtain a neighboring objective function values. Select the largest neighboring objective function value from the a neighboring objective function values ​​to obtain the largest neighboring objective function value.

[0066] If the maximum neighboring objective function value is greater than the current objective function value, then the neighboring vertex corresponding to the maximum neighboring objective function value is taken as the current vertex, and the maximum neighboring objective function value is taken as the current objective function value. The process of identifying a neighboring vertices from the feasible region of the linear programming based on the current vertex, calculating the linear objective function value of each of the a neighboring vertices, obtaining a neighboring objective function values, and selecting the maximum neighboring objective function value from the a neighboring objective function values ​​is repeated until the maximum neighboring objective function value is no greater than the current objective function value, at which point the current objective function value is taken as the maximum objective function value.

[0067] To achieve the above objectives, the present invention also provides an agricultural resource allocation system based on global optimization, comprising:

[0068] The data acquisition and partitioning module is used to receive agricultural resource allocation instructions and determine the agricultural resource allocation system and target farmland area based on the instructions. The agricultural resource allocation system includes a data acquisition module, an optimization model construction module, and a global optimization solution module. The target farmland area is divided into 'a' farmland partitions. The data acquisition module collects data from the 'a' farmland partitions to obtain 'a' multi-source agricultural datasets. Each multi-source agricultural dataset contains 'm' multi-source agricultural data points labeled with the collection time. Each multi-source agricultural data point includes: crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planting area, and available land area.

[0069] The data preprocessing module is used to preprocess a multi-source agricultural datasets to obtain a effective agricultural datasets, wherein the effective agricultural datasets, the multi-source agricultural datasets and the farmland partitions correspond one-to-one.

[0070] An optimization model construction module is used to construct a fractional structure optimization model based on the a effective agricultural datasets through an optimization model construction module. The fractional structure optimization model includes a basic objective function and a joint constraint system. Output weight coefficients and consumption weight coefficients are set. The basic objective function is weighted by the output weight coefficients and consumption weight coefficients to obtain a weighted objective function. The weighted objective function and the joint constraint system are combined to obtain a single-objective solution model.

[0071] The global optimization solution module is used to optimize the single-objective solution model through the global optimization solution module to obtain the globally optimal configuration scheme and complete the agricultural resource allocation based on global optimization.

[0072] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0073] Memory, storing at least one instruction;

[0074] The processor executes the instructions stored in the memory to implement the above-described agricultural resource allocation method based on global optimization.

[0075] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the above-described method for agricultural resource allocation based on global optimization.

[0076] To address the problems described in the background section, this invention first receives an agricultural resource allocation instruction and determines an agricultural resource allocation system and a target farmland area based on the instruction. The agricultural resource allocation system includes a data acquisition module, an optimization model construction module, and a global optimization solution module. Then, this invention divides the target farmland area into *a* farmland partitions and collects data from these partitions using the data acquisition module, obtaining *a* multi-source agricultural datasets. Each multi-source agricultural dataset contains *m* multi-source agricultural data points labeled with their collection times, and each multi-source agricultural data point includes: crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planted area, and available land. Using area, it is evident that this invention considers the problem of resource misallocation caused by the traditional method of allocating resources equally across the entire farmland, neglecting the differences in soil fertility, irrigation conditions, production capacity, and resource consumption levels among different farmland zones. Therefore, by collecting multi-source agricultural data by zone, the production characteristics and resource consumption characteristics of each farmland zone are scientifically quantified, providing data support for optimizing agricultural resource allocation. Secondly, this invention preprocesses *a* multi-source agricultural datasets to obtain *a* effective agricultural datasets, where each effective agricultural dataset, multi-source agricultural dataset, and farmland zone corresponds one-to-one. This invention recognizes that the original multi-source agricultural data contains large numerical differences between data of different dimensions, and directly using it for constructing a fractional structure optimization model would lead to calculation errors. To address the issues of errors and unreliable solutions, this invention employs feature vector transformation and normalization to obtain standardized, effective agricultural datasets. This ensures the accuracy and reliability of the subsequent construction of the fractional structure optimization model. Subsequently, the invention constructs a fractional structure optimization model based on the aforementioned *a* effective agricultural datasets using an optimization model construction module. This model includes a basic objective function and a joint constraint system. It is evident that this invention considers the limitation of traditional optimization models, which cannot simultaneously achieve both maximizing agricultural output and minimizing resource consumption, thus only achieving optimization of a single indicator. Therefore, by constructing a fractional structure optimization model, the unit resource output rate is used as the core optimization objective, while simultaneously combining single-partition constraints and global total constraints to achieve a complete optimization. Covering all objective constraints of agricultural production, this invention then sets output weight coefficients and consumption weight coefficients, and weights the basic objective function using these coefficients to obtain a weighted objective function. Combining the weighted objective function with the joint constraint system, a single-objective solution model is obtained. It is evident that this invention considers the differences in management objectives under different agricultural production scenarios. Some scenarios require prioritizing grain yield, while others require prioritizing resource consumption. A fixed basic objective function cannot flexibly adapt to different needs. Therefore, by introducing output weight coefficients and consumption weight coefficients, flexible adjustment of output and consumption priorities is achieved, ensuring that the optimization results align with different production management objectives. Finally...This invention optimizes the single-objective solution model using a global optimization module to obtain a globally optimal configuration scheme, thus achieving agricultural resource allocation based on global optimization. It is evident that this invention addresses the problem that traditional empirical configuration and simple linear programming are prone to getting trapped in local optima, failing to maximize resource utilization efficiency across the entire target farmland area. Therefore, it employs a global optimization algorithm based on the Charnes-Cooper transform and the simplex method to obtain a globally optimal configuration scheme while satisfying all constraints, achieving a scientific and rational allocation of agricultural resources. Therefore, this invention can eliminate resource mismatch problems in agricultural resource allocation, adapt to multi-dimensional resource constraint scenarios of large-scale, contiguous farmland, and ensure the maximization of agricultural resource utilization efficiency. Attached Figure Description

[0077] Figure 1 A flowchart illustrating an embodiment of the agricultural resource allocation method based on global optimization provided by the present invention;

[0078] Figure 2 A functional block diagram of an agricultural resource allocation system based on global optimization provided in an embodiment of the present invention;

[0079] Figure 3 This is a schematic diagram of the structure of an electronic device for implementing the global optimization-based agricultural resource allocation method according to an embodiment of the present invention.

[0080] Explanation of reference numerals in the attached figures:

[0081] 10. Electronic device; 11. Processor; 12. Memory; 13. Bus.

[0082] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0083] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0084] This application provides a method for agricultural resource allocation based on global optimization. The executing entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for agricultural resource allocation based on global optimization can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0085] Reference Figure 1The diagram shown is a flowchart illustrating an agricultural resource allocation method based on global optimization according to an embodiment of the present invention. In this embodiment, the agricultural resource allocation method based on global optimization includes:

[0086] S1. Receive agricultural resource allocation instructions, and determine the agricultural resource allocation system and target farmland area based on the agricultural resource allocation instructions. The agricultural resource allocation system includes: a data acquisition module, an optimization model construction module, and a global optimization solution module.

[0087] It should be understood that the agricultural resource allocation instruction is issued by personnel who wish to optimize the allocation of agricultural resources in a target farmland area. The agricultural resource allocation system is a collection of hardware and software used for agricultural data acquisition, model building, and optimization solutions, including a data acquisition module, an optimization model building module, and a global optimization solution module.

[0088] Understandably, the data acquisition module is the front-end sensing unit of the agricultural resource allocation system, used to collect multi-source agricultural data from the target farmland area. Optionally, this can be achieved using a multi-source data acquisition terminal integrating soil moisture sensors, crop growth monitors, irrigation flow meters, and agricultural machinery operation recorders. The optimization model construction module is a model construction unit used to construct a single-objective solution model. Optionally, this can be achieved using an edge computing gateway equipped with an ARM Cortex-A72 processor. The global optimization solution module is used to mathematically solve the constructed single-objective solution model to obtain the globally optimal configuration scheme. For example, this can be achieved using a cloud GPU server cluster. The target farmland area is the target geographic spatial range for agricultural resource allocation optimization, for example, 1200 mu of contiguous wheat-growing farmland in a county in northern my country or 800 mu of rice-growing base in a town in southern China as the target farmland area.

[0089] S2. Divide the target farmland area into a farmland sub-regions. Collect data from the a farmland sub-regions using the data acquisition module to obtain a multi-source agricultural datasets. Each multi-source agricultural dataset contains m multi-source agricultural data points labeled with the collection time. Each multi-source agricultural data point includes: crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planting area, and land usable area.

[0090] Understandably, the farmland zoning refers to dividing the target farmland area into multiple homogeneous production units based on soil fertility. The collection time refers to the specific moment when multi-source agricultural data is collected, used to identify the temporal information of the multi-source agricultural data, ensuring that the temporal changes in agricultural production can be accurately reproduced during subsequent preprocessing and model building. The multi-source agricultural data is a quantitative data set characterizing the agricultural production status of a single farmland zoning at a specific collection time, integrating multi-dimensional information such as crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planted area, and available land area, providing basic data support for subsequent model building.

[0091] For example, assume the target farmland area is a contiguous wheat-growing area of ​​1200 mu in a county in northern my country, divided into 3 farmland zones according to soil fertility, i.e., a=3. The total data collection period is from January 2024 to December 2024, with data collected once a month on the 15th, i.e., m=12. Each multi-source agricultural data includes: crop yield (unit: tons), crop output conversion factor (with wheat as the base crop, valued at 1.0), farmland irrigation water consumption (unit: cubic meters), production resource consumption factor (unit: standard resource units / mu, including converted values ​​of seeds, fertilizers, pesticides, agricultural machinery, etc.), actual planted area (unit: mu), and land usable area (unit: mu). Here, we take only one farmland zone as an example: On June 15, 2024, data of a certain farmland zone was collected, and the multi-source agricultural data of the sample farmland zone were as follows: crop yield 104 tons, crop output conversion factor 1.0, farmland irrigation water consumption 7600 cubic meters, production resource consumption factor 28 standard resource units / mu, actual planting area 200 mu, and land usable area 200 mu.

[0092] It should be understood that the crop yield refers to the actual total crop harvest of a single farmland zone within the total data collection period, expressed in tons, and is used to characterize the production capacity of the farmland zone. The crop output conversion factor is a coefficient used to convert the yields of different types of crops to a unified benchmark crop yield. This is used to eliminate yield differences between different crop types and to achieve horizontal comparison of output levels across different farmland zones. For example, when wheat is used as the benchmark crop, the crop output conversion factor for wheat is 1.0, and the crop output conversion factor for corn is 1.1. The farmland irrigation water consumption refers to the actual water resource consumption for crop irrigation in a single farmland zone within the total data collection period, expressed in cubic meters, and is used to characterize the water resource consumption level of the farmland zone. The production resource consumption coefficient is the converted total amount of various production resources such as seeds, fertilizers, pesticides, and agricultural machinery consumed per unit planting area in a single farmland zone within the total data collection period, expressed in standard resource units per mu (unit of land area), and is used to comprehensively characterize the non-water resource production resource consumption level of the farmland zone. The actual planted area is the land area of ​​a single farmland zone actually planted with crops during the total data collection period, measured in mu (a Chinese unit of area, approximately 0.165 acres), used to characterize the actual production scale of the farmland zone. The available land area is the maximum land area within a single farmland zone that can be used for agricultural production, measured in mu, used to limit the upper limit of the planting scale of the farmland zone.

[0093] It should be clarified that the collection period is the time interval between two consecutive multi-source agricultural data collections, used to ensure that the collected multi-source agricultural data can completely and continuously reflect the changing patterns of agricultural production status in farmland zones. The total data collection duration is the entire time span from the start of the first multi-source agricultural data collection to the end of the last multi-source agricultural data collection. The unit planting area is one acre of planted land, serving as the benchmark unit for calculating agricultural production resource consumption.

[0094] S3. Preprocess the a multi-source agricultural datasets to obtain a effective agricultural datasets, where the effective agricultural datasets and the multi-source agricultural datasets correspond one-to-one with the farmland partitions.

[0095] Understandably, the effective agricultural dataset is a standardized and time-series effective agricultural data set obtained after preprocessing, which eliminates the dimensional differences and abnormal fluctuations in multi-source agricultural data, and provides a unified and reliable data foundation for the subsequent construction of farmland partition nodes.

[0096] Specifically, the preprocessing of a multi-source agricultural datasets to obtain a effective agricultural datasets includes:

[0097] Extract multi-source agricultural datasets sequentially from 'a' multi-source agricultural datasets, and perform the following operations on each multi-source agricultural dataset:

[0098] The multi-source agricultural dataset is converted into m feature vectors. The m feature vectors are then normalized to obtain m normalized feature vectors. Each normalized feature vector corresponds one-to-one with the multi-source agricultural data, and each feature vector includes k dimensional indicators.

[0099] For each of the m normalized feature vectors, perform the following operation:

[0100] Calculate the Euclidean distance between the normalized feature vector and the preset baseline feature vector to obtain the vector distance;

[0101] If the vector distance is not greater than the preset distance threshold, the multi-source agricultural data corresponding to the normalized feature vector is confirmed as valid agricultural data, and the valid agricultural data is summarized to obtain a valid agricultural dataset.

[0102] By summing up the valid agricultural datasets, we obtain a valid agricultural datasets.

[0103] It should be understood that the feature vector is a one-dimensional numerical vector formed by arranging crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planting area, and usable land area from multi-source agricultural data in a fixed order. This vector is used to convert unstructured multi-source agricultural data into a structured data form that can be mathematically calculated. The dimension of each feature vector is consistent with the number of indicators contained in a single multi-source agricultural data set. The k-dimensional indicators are used to characterize the values ​​of various agricultural indicators in the multi-source agricultural data. The normalized feature vector is a standardized vector obtained by reorganizing all normalized data sets according to the dimensional indicators of the feature vector. This vector is used to eliminate the differences in dimensions and numerical ranges between different dimensional indicators, providing a unified quantitative basis for subsequent vector distance calculations.

[0104] For example, in this embodiment of the invention, there is multi-source agricultural data including six indicators: crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planting area, and land usable area. Each multi-source agricultural data is converted into a feature vector, and each feature vector contains six dimensional indicators, i.e., k equals 6. The six dimensional indicators are A, B, C, D, E, and F, where A, B, C, D, E, and F correspond to crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planting area, and land usable area, respectively. Taking three feature vectors as an example, let's define them as (A1, B1, C1, D1, E1, F1), (A2, B2, C2, D2, E2, F2), and (A3, B3, C3, D3, E3, F3). Divide these three feature vectors into six data groups of the same dimension based on their dimensionality: (A1, A2, A3), (B1, B2, B3), (C1, C2, C3), (D1, D2, D3), (E1, E2, E3), and (F1, F2, F3). Normalize the data group (A1, A2, A3) to obtain... Similarly, normalize the remaining 5 data groups of the same dimension to obtain 5 normalized data groups: (b1, b2, b3), (c1, c2, c3), (d1, d2, d3), (e1, e2, e3) and (f1, f2, f3). Reorganize the 6 normalized data groups according to the dimension index of the feature vector to obtain 3 normalized feature vectors: (a1, b1, c1, d1, e1, f1), (a2, b2, c2, d2, e2, f2) and (a3, b3, c3, d3, e3, f3).

[0105] It should be clarified that the benchmark feature vector is a reference vector used to determine whether the current multi-source agricultural data is normal. The benchmark feature vector is calculated from multi-source agricultural data of historical normal agricultural production in the target farmland area. The vector distance is a quantitative indicator used to characterize the similarity between the feature vector and the benchmark feature vector. The smaller the vector distance, the higher the similarity between the current multi-source agricultural data and the historical normal multi-source agricultural data, and the stronger the reliability of the current multi-source agricultural data. The calculation process of Euclidean distance (vector distance) is existing technology and will not be described in detail here.

[0106] It should be explained that the distance threshold is a manually set critical value used to filter out valid agricultural data. The distance threshold is set based on the agricultural production fluctuation characteristics of the target farmland area. The agricultural production fluctuation characteristics refer to the fluctuation range of core production indicators such as crop yield, farmland irrigation water consumption, and production resource consumption coefficient relative to the historical average level in the target farmland area under the combined influence of natural environmental factors and human management factors during production activities. For example, for target farmland areas with weak agricultural production volatility, such as contiguous high-standard farmland with flat terrain, uniform soil fertility, and stable irrigation conditions, the fluctuation range of core production indicators is less than 10%, and the distance threshold is 0.1 to 0.2. For target farmland areas with moderate agricultural production volatility, such as ordinary plain farmland and hilly farmland with good irrigation conditions, the interannual fluctuation range of core production indicators is 10% to 25%, and the distance threshold is 0.2 to 0.35. For target farmland areas with strong agricultural production volatility, such as mountainous and hilly farmland, rain-fed agricultural areas, and farmland susceptible to extreme weather, the interannual fluctuation range of core production indicators is greater than 25%, and the distance threshold is 0.35 to 0.5. The effective agricultural data refers to multi-source agricultural data that meets the requirements for similarity to historical normal multi-source agricultural data and can truly reflect the actual production status of farmland zones.

[0107] S4. Optimize the model to build a module, and construct a fractional structure optimization model based on the a effective agricultural datasets, wherein the fractional structure optimization model includes a basic objective function and a joint constraint system.

[0108] Understandably, the fractional structure optimization model is a mathematical model used for optimizing agricultural resource allocation, capable of simultaneously quantifying the comprehensive output and comprehensive consumption of agricultural production. The joint constraint system is a mathematical expression of various objective constraints in the agricultural production process, used to limit the range of values ​​for all feasible planned planting areas.

[0109] Specifically, the step of constructing a fractional structure optimization model based on the a valid agricultural datasets by optimizing the model construction module includes:

[0110] Extract the valid agricultural datasets sequentially from the 'a' valid agricultural datasets, and perform the following operation on each valid agricultural dataset:

[0111] Extract n valid agricultural data points sequentially from the valid agricultural dataset;

[0112] Based on n valid agricultural data, calculate the average crop yield per unit area, average output conversion factor, average irrigation water consumption, average resource consumption factor, and average available land area of ​​the corresponding farmland zone, and then correlate them to obtain farmland zone nodes;

[0113] Summarize the farmland partition nodes to obtain a farmland partition nodes;

[0114] Set the planned planting area for each of the a farmland zones to obtain a planned planting area. Based on the planned planting area and the a farmland zone nodes, construct a joint constraint system through the optimization model construction module.

[0115] A basic objective function is pre-constructed, and a fractional structure optimization model is formed by the joint constraint system and the basic objective function.

[0116] It should be understood that the farmland zoning node is a structured data unit formed by associating the average crop yield per unit area, average output conversion factor, average irrigation water consumption, average resource consumption factor, and average available land area of ​​a single farmland zoning area. Each farmland zoning node corresponds one-to-one with a farmland zoning area, fully encompassing the production capacity and resource consumption characteristics of that zoning area. The planned planting area is the core decision variable of the basic objective function; each farmland zoning area corresponds to one planned planting area, representing the final crop planting scale allocated to that zoning area. The basic objective function is a component of the fractional structure optimization model, used to establish a mathematical mapping relationship between the planned planting area and the unit resource output rate. The expression of the basic objective function is as follows:

[0117] ;

[0118] in, Indicates the output rate per unit of resources. Indicates the first Average crop yield per unit area in each farmland zone Indicates the first Average output conversion factor for each farmland zone Indicates the first The planned planting area for each farmland zone. Indicates the first The average resource consumption coefficient of each farmland zone This represents the preset water resource consumption factor. Indicates the first Average irrigation water consumption per farmland zone Indicates shared ownership Each farmland zone.

[0119] It should be explained that the process of setting the planned planting area is as follows: the initial setting value does not exceed the average available land area of ​​the corresponding farmland zone. In the subsequent optimization solution process, the planned planting area will be continuously adjusted within the constraints of a single zone, and finally the globally optimal planned planting area will be obtained. The unit resource output rate is the final optimization target of this invention, used to comprehensively characterize the utilization efficiency of agricultural resources. The higher the unit resource output rate, the more agricultural output is obtained under the same resource input, and the better the agricultural resource allocation scheme. The water resource consumption factor is artificially set to convert the unit irrigation water volume into an equivalent standard resource consumption, realizing the unified quantitative calculation of water consumption and the consumption of other production resources such as seeds, fertilizers, and pesticides. The setting of the water resource consumption factor is jointly determined by the water scarcity of the target farmland area, the local agricultural water-saving policy requirements, and the agricultural production resource structure. For example, when water resources are extremely scarce in the target farmland area and water conservation requirements are high, the value of the water resource consumption factor can be increased to guide the optimization scheme towards water conservation. When water resources are relatively abundant in the target farmland area, the value of the water resource consumption factor can be appropriately reduced to focus on controlling the consumption of other production resources.

[0120] Specifically, the step of calculating the average crop yield per unit area, average output conversion factor, average irrigation water consumption, average resource consumption factor, and average available land area of ​​the corresponding farmland zone based on n valid agricultural data includes:

[0121] From n valid agricultural data, we extract and summarize crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planting area and land usable area to obtain n crop yields, n crop output conversion factors, n farmland irrigation water consumption, n production resource consumption factor, n actual planting area and n land usable area.

[0122] Calculate the average output conversion factor for each of the n crops and the average usable land area for each of the n crops to obtain the average output conversion factor and the average usable land area.

[0123] The crop yields are extracted sequentially from n crop yields. Using the crop yields, the corresponding actual planting areas are retrieved from n actual planting areas. The ratio of crop yield to actual planting area is calculated to obtain the crop yield per unit area. The crop yields per unit area are then summarized to obtain n crop yields per unit area.

[0124] Calculate the average crop yield per unit area of ​​n units to obtain the average crop yield per unit area;

[0125] The average irrigation water consumption and average resource consumption coefficient are calculated based on n farmland irrigation water consumptions, n production resource consumption coefficients, and n actual planting areas.

[0126] It should be clarified that the average output conversion factor is used to characterize the crop output conversion benchmark of the farmland zone, eliminating the impact of crop type changes at different collection times on output calculation. The average available land area is used to characterize the maximum land area available for agricultural production in the farmland zone. The average crop yield per unit area is used to characterize the average unit land productivity of the farmland zone and is a key parameter for calculating the total output of the target farmland area. The crop yield per unit area is used to eliminate the impact of differences in planting area between different farmland zones on yield comparison. The average irrigation water consumption is the arithmetic mean of the irrigation water consumption per unit area of ​​farmland corresponding to all valid agricultural data within the total data collection period for a single farmland zone, used to characterize the average water resource consumption level per unit planting area of ​​the farmland zone. The average resource consumption coefficient is the arithmetic mean of the production resource consumption coefficients per unit area corresponding to all valid agricultural data within the total data collection period for a single farmland zone, used to characterize the average non-water resource production resource consumption level per unit planting area of ​​the farmland zone.

[0127] Specifically, the construction of a joint constraint system based on a planned planting area and a farmland zoning nodes through an optimization model construction module includes:

[0128] Based on a planned planting area and a farmland partition nodes, a single partition constraint is constructed through an optimization model construction module, wherein the single partition constraint is as follows:

[0129] ;

[0130] in, Indicates the first The planned planting area for each farmland zone. Indicates the first The average usable land area of ​​each farmland zone Indicates shared ownership One farmland zone;

[0131] From each of the a farmland partition nodes, extract and summarize the average irrigation water consumption, average resource consumption coefficient, and average available land area to obtain a average irrigation water consumption, a average resource consumption coefficient, and a average available land area.

[0132] Calculate the sum of a planned planting areas and the sum of a average available land areas to obtain the total planned planting area and total available land area of ​​the target farmland area.

[0133] The total irrigation water consumption is calculated based on 'a' average irrigation water consumption and 'a' planned planting area, as shown below:

[0134] ;

[0135] in, This represents the total irrigation water consumption. Indicates the first Average irrigation water consumption per farmland zone Indicates the first The planned planting area for each farmland zone. Indicates shared ownership One farmland zone;

[0136] Calculate the total resource consumption based on a average resource consumption coefficient and a planned planting area;

[0137] Based on the total planned planting area, total available land area, total irrigation water consumption, and total resource consumption, a global total constraint is constructed through an optimization model construction module. The global total constraint is as follows:

[0138] ;

[0139] in, This represents the total planned planting area. This represents the total available land area. This represents the total irrigation water consumption. This indicates the preset irrigation water threshold. This represents the total resource consumption. This indicates the preset resource consumption threshold. Indicates the total quantity identifier. Indicates the default value identifier;

[0140] By summarizing the single-partition constraints and the global total constraints, a joint constraint system is obtained.

[0141] Understandably, the single-zone constraint is a constraint on the upper limit of planting scale for a single farmland zone, used to ensure that the planting area of ​​a single zone does not exceed the corresponding usable land area. The total planned planting area is the sum of 'a' planned planting areas, used to characterize the overall production scale of the target farmland area. The total usable land area is the sum of 'a' average usable land areas, used to limit the maximum production scale of the target farmland area. The total irrigation water consumption is used to characterize the overall water resource consumption level of the target farmland area. The total resource consumption is used to characterize the overall production material consumption level of the target farmland area; the calculation process for the total resource consumption is the same as that for the total irrigation water consumption, and will not be repeated here. The global total constraint is a constraint on the upper limit of resource consumption for the entire target farmland area, used to ensure that the overall resource consumption does not exceed the actual carrying capacity. The irrigation water threshold is a manually set maximum total irrigation water consumption allowed for the target farmland area. The irrigation water threshold is set based on the agricultural water quota allocated to the target farmland area by the local water conservancy department, while also considering the rainfall of the target farmland area. The resource consumption threshold is the maximum total production resource consumption allowed for the target farmland area, which is set by human intervention. The resource consumption threshold is set based on the agricultural production input budget of the target farmland area, and in combination with the production material supply capacity of the target farmland area.

[0142] It should be clarified that the process of single-zone constraint is as follows: The planned planting area and the corresponding average available land area of ​​each farmland zone are extracted sequentially, and it is determined whether the planned planting area is not less than 0 and not greater than the average available land area of ​​the farmland zone. If this condition is met, the planned planting area of ​​the farmland zone meets the requirements; otherwise, the value of the planned planting area needs to be readjusted. The process of global total constraint is as follows: If the total planned planting area is not greater than the total available land area, the total irrigation water consumption is not greater than the irrigation water threshold, and the total resource consumption is not greater than the resource consumption threshold, then the current planned planting area allocation scheme is considered to meet the global resource carrying capacity requirements; otherwise, the value of the planned planting area needs to be readjusted.

[0143] For example, the local water conservancy department allocates 45,000 cubic meters of agricultural water to 1,200 mu of wheat-growing farmland. Therefore, the irrigation water threshold is set at 45,000 cubic meters. The agricultural production input budget corresponds to 35,000 standard resource units, so the resource consumption threshold is set at 35,000 standard resource units. In this embodiment of the invention, the sample farmland partition nodes corresponding to the sample farmland partition mentioned above are: average crop yield of 0.52 tons / mu, average output conversion factor of 1.0, average irrigation water consumption of 38 cubic meters / mu, average resource consumption factor of 28 standard resource units / mu, and average available land area of ​​200 mu. The planned planting area of ​​the sample farmland partition is set to 200 mu, which satisfies the single partition constraint. The calculated irrigation water consumption of the sample farmland partition is 7,600 cubic meters, and the resource consumption is 5,600 standard resource units, which meets the overall resource carrying capacity requirements of the target farmland area.

[0144] S5. Set the output weight coefficient and the consumption weight coefficient, and apply the weighting to the basic objective function to obtain the weighted objective function. Combine the weighted objective function and the joint constraint system to obtain the single-objective solution model.

[0145] Understandingly, the output weight coefficient is a parameter used to adjust the importance of the numerator in the basic objective function, representing the degree of importance attached to agricultural output. By adjusting the value of the output weight coefficient, the weight proportion of the numerator in the basic objective function can be changed. The output weight coefficient is set according to the agricultural production management objectives of the target farmland area. For example, when prioritizing grain production is the main objective, the value of the output weight coefficient can be increased; when considering resource utilization efficiency is the main objective, the value of the output weight coefficient can be appropriately decreased. The consumption weight coefficient is a parameter used to adjust the importance of the denominator in the basic objective function, representing the degree of importance attached to resource consumption control. By adjusting the value of the consumption weight coefficient, the weight proportion of the consumption term in the basic objective function can be changed. The consumption weight coefficient is set according to the agricultural production management objectives of the target farmland area. For example, when prioritizing resource consumption control and improving resource utilization efficiency is the main objective, the value of the consumption weight coefficient can be increased; when prioritizing grain production is the main objective, the value of the consumption weight coefficient can be appropriately decreased.

[0146] It should be understood that the weighting process involves multiplying the output weight coefficient by the numerator of the basic objective function and the consumption weight coefficient by the denominator of the basic objective function, thereby adjusting the weights of the numerator and denominator of the basic objective function. This quantifies the optimization objective into the basic objective function, ensuring that the globally optimal configuration scheme obtained from subsequent optimization solutions meets the requirements. The weighted objective function is the function obtained after weighting the basic objective function. It is used to calculate the weighted unit resource output rate for different planned planting areas, providing a basis for subsequent optimization solutions. The weighted unit resource output rate is a quantitative indicator that comprehensively considers the priority of agricultural output and the priority of resource consumption. The higher the value of this indicator, the better the overall resource utilization effect. The expression of the weighted objective function is as follows:

[0147] ;

[0148] in, This represents the weighted unit resource output rate. This represents the output weighting coefficient. This represents the consumption weighting coefficient. Indicates the first Average crop yield per unit area in each farmland zone Indicates the first Average output conversion factor for each farmland zone Indicates the first The planned planting area for each farmland zone. Indicates the first The average resource consumption coefficient of each farmland zone This represents the preset water resource consumption factor. Indicates the first Average irrigation water consumption per farmland zone Indicates shared ownership Each farmland zone.

[0149] It should be clarified that the single-objective solution model is a complete mathematical model composed of a weighted objective function and a joint constraint system. It includes both the optimization objective and the constraints of all feasible planting areas to be planned. This model provides a directly solvable model framework for the subsequent global optimization solution module, ensuring that the optimization process is within the feasible range of the joint constraint system.

[0150] S6. The single-objective solution model is optimized and solved by the global optimization solution module to obtain the globally optimal configuration scheme, thus completing the agricultural resource allocation based on global optimization.

[0151] It is understood that the global optimal configuration scheme is the optimal agricultural resource allocation result obtained through optimization and solution. It includes the optimal planned planting area of ​​all farmland zones and the optimal unit resource output rate of the target farmland area as a whole. Under the premise of meeting all objective constraints such as land, water resources, and production materials, it can maximize the efficiency of agricultural resource utilization.

[0152] Specifically, the step of optimizing the single-objective solution model using a global optimization solution module to obtain a globally optimal configuration scheme includes:

[0153] Applying the Charnes-Cooper transform to the weighted objective function yields a linear objective function;

[0154] The joint constraint system is synchronously converted into a linear constraint system, and a standard linear programming model is obtained based on the linear objective function and the linear constraint system.

[0155] The feasible region of a linear programming problem is determined according to a standard linear programming model, wherein the feasible region of a linear programming problem includes multiple vertices;

[0156] Calculate the maximum objective function value corresponding to the feasible region of the linear programming problem, and calculate the optimal planting area to be planned based on the maximum objective function value;

[0157] Substitute the optimal planned planting area into the basic objective function to calculate the optimal unit resource output rate of the target farmland area.

[0158] By summarizing the optimal planned planting area and the optimal unit resource output rate, the globally optimal allocation scheme is obtained.

[0159] Understandably, the linear objective function is a linear form function obtained by performing a Charnes-Cooper transformation on the weighted objective function. All its variables are linear terms, conforming to the standard form of linear programming, and can be directly solved using a linear programming algorithm. The linear programming algorithm is existing technology, and its specific process will not be elaborated here. The expression of the linear objective function is as follows:

[0160] ;

[0161] in, This represents the value of the linear objective function. This represents the output weighting coefficient. Indicates the first Average crop yield per unit area in each farmland zone Indicates the first Average output conversion factor for each farmland zone Indicates shared ownership Each farmland zone Indicates the first The transformed decision variables for each farmland zoning.

[0162] It should be understood that the linear constraint system is a set of constraints obtained by simultaneously converting all nonlinear constraints in the joint constraint system into linear form through the Charnes-Cooper transformation. It includes linear single-partition constraints, linear total constraints, and linear normalization constraints, completely preserving all the restrictions of the original constraints and ensuring that the solution obtained still satisfies all objective production conditions. The standard linear programming model is a complete mathematical model composed of a linear objective function and a linear constraint system, used to calculate all solutions that satisfy the linear constraint system. The feasible region of the linear programming is the set of all solutions in the standard linear programming model that satisfy the linear constraint system. Geometrically, it is represented as a convex polyhedron. All possible solutions lie inside or on the boundary of this convex polyhedron. According to the fundamental theorem of linear programming, the optimal solution of the linear programming must appear at the vertices of this convex polyhedron. The vertex is a point on the convex polyhedron corresponding to the feasible region of the linear programming. Each vertex corresponds to a unique feasible solution and is the basic unit of iteration for the linear programming algorithm. The maximum objective function value is the linear objective function value corresponding to the optimal vertex obtained after the iterative convergence of the linear programming algorithm. Numerically, it is equivalent to the maximum value of the weighted objective function, representing the highest weighted unit resource output rate achievable among all feasible solutions. The linear objective function value is the calculation result of the linear objective function at the vertex, used to measure the optimization effect of the solution corresponding to that vertex.

[0163] It should be clarified that the optimal planned planting area is obtained by inversely transforming the transformed decision variables corresponding to the maximum objective function value, representing the final planned planting area allocated to each farmland partition, and is a core component of the globally optimal allocation scheme. The optimal unit resource output rate is the final optimization index calculated by substituting the optimal planned planting area into the basic objective function. It represents the agricultural output level that a target farmland area can obtain per unit of comprehensive resource input under the optimal resource allocation scheme, and is used to measure the merits of the globally optimal allocation scheme. The globally optimal allocation scheme is the final result obtained by combining the optimal planned planting area and the optimal unit resource output rate, and can maximize the efficiency of agricultural resource allocation and utilization in the target farmland area while satisfying all realistic resource constraints.

[0164] Specifically, the step of performing a Charnes-Cooper transform on the weighted objective function to obtain a linear objective function includes:

[0165] A global transformation variable and *a* transformed decision variables are introduced, wherein each transformed decision variable corresponds one-to-one with a farmland zone. The global transformation variable and *a* transformed decision variables are shown below:

[0166] ;

[0167] in, Represents the global transformation variable. This represents the consumption weighting coefficient. Indicates the first The average resource consumption coefficient of each farmland zone This represents the preset water resource consumption factor. Indicates the first Average irrigation water consumption per farmland zone Indicates the first The planned planting area for each farmland zone. Indicates the first Decision variables after transformation of each farmland zoning Indicates shared ownership One farmland zone;

[0168] Substituting the global transformation variable and the a transformed decision variables into the weighted objective function yields a linear objective function.

[0169] It should be explained that the global transformation variable is an auxiliary variable introduced in the Charnes-Cooper transformation, used to convert the weighted objective function into a linear form. The transformed decision variable is an auxiliary variable introduced in the Charnes-Cooper transformation, used to linearize the planned planting area.

[0170] Specifically, the step of synchronously converting the joint constraint system into a linear constraint system includes:

[0171] Extract single-partition constraints and global total constraints from the joint constraint system;

[0172] Based on *a* transformed decision variables, global transformation variables, and *a* average available land areas, the single-zone constraint is transformed into a linear single-zone constraint, as shown below:

[0173] ;

[0174] in, Indicates the first Decision variables after transformation of each farmland zoning Represents the global transformation variable. Indicates the first The average usable land area of ​​each farmland zone Indicates shared ownership One farmland zone;

[0175] The linear total planting area, linear total irrigation water consumption, and linear total resource consumption are calculated based on a transformed decision variables, global transformation variables, a average irrigation water consumption, and a average resource consumption coefficient.

[0176] The global total constraint is transformed into a linear total constraint based on the linear total planting area, linear total irrigation water consumption, and linear total resource consumption. The linear total constraint is as follows:

[0177] ;

[0178] in, This represents the total linear planting area. This represents the total available land area. Represents the global transformation variable. This represents the total linear irrigation water consumption. This indicates the preset irrigation water threshold. This represents the total linear resource consumption. This indicates the preset resource consumption threshold. Indicates a linear total quantity identifier. Indicates the total quantity identifier. Indicates the default value identifier;

[0179] A linear normalization constraint is constructed based on the total linear resource consumption and the consumption weight coefficient, as shown below:

[0180] ;

[0181] in, This represents the consumption weighting coefficient. This represents the total linear resource consumption;

[0182] By summarizing linear single-partition constraints, linear total constraints, and linear normalization constraints, a system of linear constraint conditions is obtained.

[0183] It should be clarified that the linear single-partition constraint is a linear form constraint obtained by transforming the single-partition constraint through Charnes-Cooper. Its constraint method is the same as that of the single-partition constraint and will not be repeated here. The linear total planted area is a linear index obtained by transforming the total planted area through Charnes-Cooper, used to characterize the total scale of land resources after the transformation. The linear total irrigation water consumption is a linear index obtained by transforming the total irrigation water consumption through Charnes-Cooper, used to characterize the total water resource consumption level after the transformation. The linear total resource consumption is a linear index obtained by transforming the total resource consumption through Charnes-Cooper, used to characterize the comprehensive total resource consumption level after the transformation. The linear total constraint is a linear constraint obtained by transforming the global total constraint through Charnes-Cooper, used to limit the overall resource consumption upper limit of the target farmland area. The constraint method of the linear total constraint is the same as that of the global total constraint and will not be repeated here. The linear normalization constraint is a constraint condition constructed to ensure the equivalence of the model before and after the Charnes-Cooper transformation, obtained by combining the transformed linear total resource consumption with the consumption weight coefficient.

[0184] Specifically, calculating the maximum objective function value corresponding to the feasible region of the linear programming includes:

[0185] Based on the linear constraint system, the current vertex is identified from multiple vertices;

[0186] The linear objective function value corresponding to the current vertex is calculated to obtain the current objective function value;

[0187] Based on the current vertex, identify a neighboring vertices from the feasible region of the linear programming. Calculate the linear objective function value of each of the a neighboring vertices to obtain a neighboring objective function values. Select the largest neighboring objective function value from the a neighboring objective function values ​​to obtain the largest neighboring objective function value.

[0188] If the maximum neighboring objective function value is greater than the current objective function value, then the neighboring vertex corresponding to the maximum neighboring objective function value is taken as the current vertex, and the maximum neighboring objective function value is taken as the current objective function value. The process of identifying a neighboring vertices from the feasible region of the linear programming based on the current vertex, calculating the linear objective function value of each of the a neighboring vertices, obtaining a neighboring objective function values, and selecting the maximum neighboring objective function value from the a neighboring objective function values ​​is repeated until the maximum neighboring objective function value is no greater than the current objective function value, at which point the current objective function value is taken as the maximum objective function value.

[0189] Understandably, the current vertex is the starting point for each iteration, selected within all vertices of the feasible region of the linear programming problem, satisfying all requirements of the linear constraint system. The current objective function value is the linear objective function value corresponding to the current vertex, used to compare with the maximum neighbor objective function value to determine the direction of iterative optimization. The neighboring vertices are those directly connected to the current vertex; each neighboring vertex corresponds to a new feasible solution and serves as a candidate calculation point in the iteration process. The neighboring objective function value is the linear objective function value corresponding to the neighboring vertex, used to evaluate the optimization level of the candidate feasible solution. The maximum neighbor objective function value is the maximum value among all the linear objective function values ​​corresponding to all neighboring vertices of a given current vertex, representing the optimal candidate solution achievable in this iteration. The criterion for stopping iteration when the maximum neighboring objective function value is not greater than the current objective function value is as follows: The iterative logic of the iterative optimization is to always move in the direction of strictly increasing linear objective function value. Each iteration starts from the current vertex and selects the neighboring vertex with the largest linear objective function value as the next iteration point. When the maximum neighboring objective function value is not greater than the current objective function value, it means that none of the neighboring vertices of the current vertex can provide a better linear objective function value. That is, there is no vertex better than the current vertex. The corresponding current objective function value is the global maximum objective function value. Continuing the iteration will not yield a better result, so the iteration process is terminated.

[0190] To address the problems described in the background section, this invention first receives an agricultural resource allocation instruction and determines an agricultural resource allocation system and a target farmland area based on the instruction. The agricultural resource allocation system includes a data acquisition module, an optimization model construction module, and a global optimization solution module. Then, this invention divides the target farmland area into *a* farmland partitions and collects data from these partitions using the data acquisition module, obtaining *a* multi-source agricultural datasets. Each multi-source agricultural dataset contains *m* multi-source agricultural data points labeled with their collection times, and each multi-source agricultural data point includes: crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planted area, and available land. Using area, it is evident that this invention considers the problem of resource misallocation caused by the traditional method of allocating resources equally across the entire farmland, neglecting the differences in soil fertility, irrigation conditions, production capacity, and resource consumption levels among different farmland zones. Therefore, by collecting multi-source agricultural data by zone, the production characteristics and resource consumption characteristics of each farmland zone are scientifically quantified, providing data support for optimizing agricultural resource allocation. Secondly, this invention preprocesses *a* multi-source agricultural datasets to obtain *a* effective agricultural datasets, where each effective agricultural dataset, multi-source agricultural dataset, and farmland zone corresponds one-to-one. This invention recognizes that the original multi-source agricultural data contains large numerical differences between data of different dimensions, and directly using it for constructing a fractional structure optimization model would lead to calculation errors. To address the issues of errors and unreliable solutions, this invention employs feature vector transformation and normalization to obtain standardized, effective agricultural datasets. This ensures the accuracy and reliability of the subsequent construction of the fractional structure optimization model. Subsequently, the invention constructs a fractional structure optimization model based on the aforementioned *a* effective agricultural datasets using an optimization model construction module. This model includes a basic objective function and a joint constraint system. It is evident that this invention considers the limitation of traditional optimization models, which cannot simultaneously achieve both maximizing agricultural output and minimizing resource consumption, thus only achieving optimization of a single indicator. Therefore, by constructing a fractional structure optimization model, the unit resource output rate is used as the core optimization objective, while simultaneously combining single-partition constraints and global total constraints to achieve a complete optimization. Covering all objective constraints of agricultural production, this invention then sets output weight coefficients and consumption weight coefficients, and weights the basic objective function using these coefficients to obtain a weighted objective function. Combining the weighted objective function with the joint constraint system, a single-objective solution model is obtained. It is evident that this invention considers the differences in management objectives under different agricultural production scenarios. Some scenarios require prioritizing grain yield, while others require prioritizing resource consumption. A fixed basic objective function cannot flexibly adapt to different needs. Therefore, by introducing output weight coefficients and consumption weight coefficients, flexible adjustment of output and consumption priorities is achieved, ensuring that the optimization results align with different production management objectives. Finally...This invention optimizes the single-objective solution model using a global optimization module to obtain a globally optimal configuration scheme, thus achieving agricultural resource allocation based on global optimization. It is evident that this invention addresses the problem that traditional empirical configuration and simple linear programming are prone to getting trapped in local optima, failing to maximize resource utilization efficiency across the entire target farmland area. Therefore, it employs a global optimization algorithm based on the Charnes-Cooper transform and the simplex method to obtain a globally optimal configuration scheme while satisfying all constraints, achieving a scientific and rational allocation of agricultural resources. Therefore, this invention can eliminate resource mismatch problems in agricultural resource allocation, adapt to multi-dimensional resource constraint scenarios of large-scale, contiguous farmland, and ensure the maximization of agricultural resource utilization efficiency.

[0191] like Figure 2 The diagram shown is a functional block diagram of an agricultural resource allocation system based on global optimization provided in an embodiment of the present invention.

[0192] The agricultural resource allocation system 100 based on global optimization described in this invention can be installed in an electronic device. Depending on the functions implemented, the agricultural resource allocation system 100 based on global optimization may include a data acquisition and partitioning module 101, a data preprocessing module 102, an optimization model construction module 103, and a global optimization solution module 104. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and which are stored in the memory of the electronic device.

[0193] The data acquisition partitioning module 101 is used to receive agricultural resource allocation instructions and determine the agricultural resource allocation system and target farmland area based on the agricultural resource allocation instructions. The agricultural resource allocation system includes: a data acquisition module, an optimization model construction module, and a global optimization solution module. The target farmland area is divided into a farmland partitions. Data is collected from the a farmland partitions by the data acquisition module to obtain a multi-source agricultural datasets. Each multi-source agricultural dataset contains m multi-source agricultural data points labeled with the collection time. Each multi-source agricultural data point includes: crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planting area, and land usable area.

[0194] The data preprocessing module 102 is used to preprocess a multi-source agricultural datasets to obtain a effective agricultural datasets, wherein the effective agricultural datasets and the multi-source agricultural datasets correspond one-to-one with farmland partitions;

[0195] The optimization model construction module 103 is used to construct a fractional structure optimization model based on the a effective agricultural datasets through the optimization model construction module. The fractional structure optimization model includes a basic objective function and a joint constraint system. Output weight coefficients and consumption weight coefficients are set. The basic objective function is weighted by the output weight coefficients and consumption weight coefficients to obtain a weighted objective function. The single objective solution model is obtained by combining the weighted objective function and the joint constraint system.

[0196] The global optimization solution module 104 is used to optimize the single-objective solution model through the global optimization solution module to obtain the globally optimal configuration scheme and complete the agricultural resource allocation based on global optimization.

[0197] In detail, the modules in the globally optimized agricultural resource allocation system 100 described in this embodiment of the invention employ the same methods as described above. Figure 1 The method uses the same technical means as the global optimization-based agricultural resource allocation method described in the article and can produce the same technical effects, so it will not be repeated here.

[0198] like Figure 3 The diagram shown is a schematic representation of an electronic device for implementing a globally optimized agricultural resource allocation method according to an embodiment of the present invention.

[0199] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as an agricultural resource allocation method program based on global optimization.

[0200] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as the portable hard drive of the electronic device 1. In other embodiments, the memory 11 can be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 includes both internal storage units and external storage devices of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as the code of a globally optimized agricultural resource allocation method program, but also to temporarily store data that has been output or will be output.

[0201] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., a globally optimized agricultural resource allocation method program) and calls data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0202] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0203] Figure 3 Only electronic devices with components are shown; those skilled in the art will understand that... Figure 3 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0204] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0205] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device 1 and other electronic devices.

[0206] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0207] The agricultural resource allocation method program based on global optimization stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When run in the processor 10, it can achieve the following:

[0208] The system receives agricultural resource allocation instructions and determines an agricultural resource allocation system and target farmland area based on the instructions. The agricultural resource allocation system includes a data acquisition module, an optimization model construction module, and a global optimization solution module.

[0209] The target farmland area is divided into a farmland partitions. Data is collected from the a farmland partitions through the data acquisition module to obtain a multi-source agricultural datasets. Each multi-source agricultural dataset contains m multi-source agricultural data points labeled with the collection time. Each multi-source agricultural data point includes: crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planting area and land usable area.

[0210] Preprocessing a multi-source agricultural datasets yields a valid agricultural datasets, where the valid agricultural datasets, multi-source agricultural datasets, and farmland partitions correspond one-to-one.

[0211] By optimizing the model construction module, a fractional structure optimization model is constructed based on the a effective agricultural datasets, wherein the fractional structure optimization model includes a basic objective function and a joint constraint system;

[0212] Set output weight coefficients and consumption weight coefficients, and apply weighting to the basic objective function using these coefficients to obtain a weighted objective function. Combine the weighted objective function with the joint constraint system to obtain a single-objective solution model.

[0213] The single-objective solution model is optimized and solved by the global optimization solution module to obtain the globally optimal configuration scheme, thus completing the agricultural resource allocation based on global optimization.

[0214] Specifically, the processor 10's implementation method for the above instructions can be found in [reference needed]. Figures 1 to 3 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0215] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0216] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0217] The system receives agricultural resource allocation instructions and determines an agricultural resource allocation system and target farmland area based on the instructions. The agricultural resource allocation system includes a data acquisition module, an optimization model construction module, and a global optimization solution module.

[0218] The target farmland area is divided into a farmland partitions. Data is collected from the a farmland partitions through the data acquisition module to obtain a multi-source agricultural datasets. Each multi-source agricultural dataset contains m multi-source agricultural data points labeled with the collection time. Each multi-source agricultural data point includes: crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planting area and land usable area.

[0219] Preprocessing a multi-source agricultural datasets yields a valid agricultural datasets, where the valid agricultural datasets, multi-source agricultural datasets, and farmland partitions correspond one-to-one.

[0220] By optimizing the model construction module, a fractional structure optimization model is constructed based on the a effective agricultural datasets, wherein the fractional structure optimization model includes a basic objective function and a joint constraint system;

[0221] Set output weight coefficients and consumption weight coefficients, and apply weighting to the basic objective function using these coefficients to obtain a weighted objective function. Combine the weighted objective function with the joint constraint system to obtain a single-objective solution model.

[0222] The single-objective solution model is optimized and solved by the global optimization solution module to obtain the globally optimal configuration scheme, thus completing the agricultural resource allocation based on global optimization.

[0223] In the embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and actual implementations may have other classification methods.

[0224] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0225] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0226] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0227] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for agricultural resource allocation based on global optimization, characterized in that, The method includes: The system receives agricultural resource allocation instructions and determines an agricultural resource allocation system and target farmland area based on the instructions. The agricultural resource allocation system includes a data acquisition module, an optimization model construction module, and a global optimization solution module. The target farmland area is divided into a farmland partitions. Data is collected from the a farmland partitions through the data acquisition module to obtain a multi-source agricultural datasets. Each multi-source agricultural dataset contains m multi-source agricultural data points labeled with the collection time. Each multi-source agricultural data point includes: crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planting area and land usable area. Preprocessing a multi-source agricultural datasets yields a valid agricultural datasets, where the valid agricultural datasets, multi-source agricultural datasets, and farmland partitions correspond one-to-one. By optimizing the model construction module, a fractional structure optimization model is constructed based on the a effective agricultural datasets, wherein the fractional structure optimization model includes a basic objective function and a joint constraint system; Set output weight coefficients and consumption weight coefficients, and apply weighting to the basic objective function using these coefficients to obtain a weighted objective function. Combine the weighted objective function with the joint constraint system to obtain a single-objective solution model. The single-objective solution model is optimized and solved by the global optimization solution module to obtain the globally optimal configuration scheme, thus completing the agricultural resource allocation based on global optimization.

2. The agricultural resource allocation method based on global optimization as described in claim 1, characterized in that, The preprocessing of *a* multi-source agricultural datasets to obtain *a* effective agricultural datasets includes: Extract multi-source agricultural datasets sequentially from 'a' multi-source agricultural datasets, and perform the following operations on each multi-source agricultural dataset: The multi-source agricultural dataset is converted into m feature vectors. The m feature vectors are then normalized to obtain m normalized feature vectors. Each normalized feature vector corresponds one-to-one with the multi-source agricultural data, and each feature vector includes k dimensional indicators. For each of the m normalized feature vectors, perform the following operation: Calculate the Euclidean distance between the normalized feature vector and the preset baseline feature vector to obtain the vector distance; If the vector distance is not greater than the preset distance threshold, the multi-source agricultural data corresponding to the normalized feature vector is confirmed as valid agricultural data, and the valid agricultural data is summarized to obtain a valid agricultural dataset. By summing up the valid agricultural datasets, we obtain a valid agricultural datasets.

3. The agricultural resource allocation method based on global optimization as described in claim 1, characterized in that, The step of constructing a module by optimizing the model, based on the a valid agricultural datasets, includes: Extract the valid agricultural datasets sequentially from the 'a' valid agricultural datasets, and perform the following operation on each valid agricultural dataset: Extract n valid agricultural data points sequentially from the valid agricultural dataset; Based on n valid agricultural data, calculate the average crop yield per unit area, average output conversion factor, average irrigation water consumption, average resource consumption factor, and average available land area of ​​the corresponding farmland zone, and then correlate them to obtain farmland zone nodes; Summarize the farmland partition nodes to obtain a farmland partition nodes; Set the planned planting area for each of the a farmland zones to obtain a planned planting area. Based on the planned planting area and the a farmland zone nodes, construct a joint constraint system through the optimization model construction module. A basic objective function is pre-constructed, and a fractional structure optimization model is formed by the joint constraint system and the basic objective function.

4. The agricultural resource allocation method based on global optimization as described in claim 3, characterized in that, The calculation of the average crop yield per unit area, average output conversion factor, average irrigation water consumption, average resource consumption factor, and average available land area for the corresponding farmland zone based on n valid agricultural data includes: From n valid agricultural data, we extract and summarize crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planting area and land usable area to obtain n crop yields, n crop output conversion factors, n farmland irrigation water consumption, n production resource consumption factor, n actual planting area and n land usable area. Calculate the average output conversion factor for each of the n crops and the average usable land area for each of the n crops to obtain the average output conversion factor and the average usable land area. The crop yields are extracted sequentially from n crop yields. Using the crop yields, the corresponding actual planting areas are retrieved from n actual planting areas. The ratio of crop yield to actual planting area is calculated to obtain the crop yield per unit area. The crop yields per unit area are then summarized to obtain n crop yields per unit area. Calculate the average crop yield per unit area of ​​n units to obtain the average crop yield per unit area; The average irrigation water consumption and average resource consumption coefficient are calculated based on n farmland irrigation water consumptions, n production resource consumption coefficients, and n actual planting areas.

5. The agricultural resource allocation method based on global optimization as described in claim 3, characterized in that, The joint constraint system is constructed by optimizing the model construction module based on a planned planting area and a farmland zoning nodes, including: Based on a planned planting area and a farmland partition nodes, single partition constraints are constructed by optimizing the model construction module; From each of the a farmland partition nodes, extract and summarize the average irrigation water consumption, average resource consumption coefficient, and average available land area to obtain a average irrigation water consumption, a average resource consumption coefficient, and a average available land area. Calculate the sum of a planned planting areas and the sum of a average available land areas to obtain the total planned planting area and total available land area of ​​the target farmland area. Calculate the total irrigation water consumption based on a average irrigation water consumption and a planned planting area; Calculate the total resource consumption based on a average resource consumption coefficient and a planned planting area; Based on the total planned planting area, total available land area, total irrigation water consumption, and total resource consumption, a global total constraint is constructed through an optimization model construction module. The global total constraint is as follows: ； in, This represents the total planned planting area. This represents the total available land area. This represents the total irrigation water consumption. This indicates the preset irrigation water threshold. This represents the total resource consumption. This indicates the preset resource consumption threshold. Indicates the total quantity identifier. Indicates the default value identifier; By summarizing the single-partition constraints and the global total constraints, a joint constraint system is obtained.

6. The agricultural resource allocation method based on global optimization as described in claim 1, characterized in that, The step of optimizing the single-objective solution model using a global optimization solution module to obtain the globally optimal configuration scheme includes: Applying the Charnes-Cooper transform to the weighted objective function yields a linear objective function; The joint constraint system is synchronously converted into a linear constraint system, and a standard linear programming model is obtained based on the linear objective function and the linear constraint system. The feasible region of a linear programming problem is determined according to a standard linear programming model, wherein the feasible region of a linear programming problem includes multiple vertices; Calculate the maximum objective function value corresponding to the feasible region of the linear programming problem, and calculate the optimal planting area to be planned based on the maximum objective function value; Substitute the optimal planned planting area into the basic objective function to calculate the optimal unit resource output rate of the target farmland area. By summarizing the optimal planned planting area and the optimal unit resource output rate, the globally optimal allocation scheme is obtained.

7. The agricultural resource allocation method based on global optimization as described in claim 6, characterized in that, The step of performing a Charnes-Cooper transform on the weighted objective function to obtain a linear objective function includes: A global transformation variable and *a* transformed decision variables are introduced, wherein each transformed decision variable corresponds one-to-one with a farmland zone. The global transformation variable and *a* transformed decision variables are shown below: ； in, Represents the global transformation variable. This represents the consumption weighting coefficient. Indicates the first The average resource consumption coefficient of each farmland zone This represents the preset water resource consumption factor. Indicates the first Average irrigation water consumption per farmland zone Indicates the first The planned planting area for each farmland zone. Indicates the first Decision variables after transformation of each farmland zoning Indicates shared ownership One farmland zone; Substituting the global transformation variable and the a transformed decision variables into the weighted objective function yields a linear objective function.

8. The agricultural resource allocation method based on global optimization as described in claim 6, characterized in that, The process of simultaneously converting the joint constraint system into a linear constraint system includes: Extract single-partition constraints and global total constraints from the joint constraint system; Based on a transformed decision variables, global transformation variables, and a average available land area, the single-zone constraint is transformed into a linear single-zone constraint. The linear total planting area, linear total irrigation water consumption, and linear total resource consumption are calculated based on a transformed decision variables, global transformation variables, a average irrigation water consumption, and a average resource consumption coefficient. The global total constraint is converted into a linear total constraint based on the linear total planting area, linear total irrigation water consumption, and linear total resource consumption. Linear normalization constraints are constructed based on the linear total resource consumption and consumption weight coefficients; By summarizing linear single-partition constraints, linear total constraints, and linear normalization constraints, a system of linear constraint conditions is obtained.

9. The agricultural resource allocation method based on global optimization as described in claim 6, characterized in that, The calculation of the maximum objective function value corresponding to the feasible region of the linear programming includes: Based on the linear constraint system, the current vertex is identified from multiple vertices; The linear objective function value corresponding to the current vertex is calculated to obtain the current objective function value; Based on the current vertex, identify a neighboring vertices from the feasible region of the linear programming. Calculate the linear objective function value of each of the a neighboring vertices to obtain a neighboring objective function values. Select the largest neighboring objective function value from the a neighboring objective function values ​​to obtain the largest neighboring objective function value. If the maximum neighboring objective function value is greater than the current objective function value, then the neighboring vertex corresponding to the maximum neighboring objective function value is taken as the current vertex, and the maximum neighboring objective function value is taken as the current objective function value. The process of identifying a neighboring vertices from the feasible region of the linear programming based on the current vertex, calculating the linear objective function value of each of the a neighboring vertices, obtaining a neighboring objective function values, and selecting the maximum neighboring objective function value from the a neighboring objective function values ​​is repeated until the maximum neighboring objective function value is no greater than the current objective function value, at which point the current objective function value is taken as the maximum objective function value.

10. An agricultural resource allocation system based on global optimization, characterized in that, The system includes: The data acquisition and partitioning module is used to receive agricultural resource allocation instructions and determine the agricultural resource allocation system and target farmland area based on the instructions. The agricultural resource allocation system includes a data acquisition module, an optimization model construction module, and a global optimization solution module. The target farmland area is divided into 'a' farmland partitions. The data acquisition module collects data from the 'a' farmland partitions to obtain 'a' multi-source agricultural datasets. Each multi-source agricultural dataset contains 'm' multi-source agricultural data points labeled with the collection time. Each multi-source agricultural data point includes: crop yield, crop output conversion factor, farmland irrigation water consumption, production resource consumption factor, actual planting area, and available land area. The data preprocessing module is used to preprocess a multi-source agricultural datasets to obtain a effective agricultural datasets, wherein the effective agricultural datasets, the multi-source agricultural datasets and the farmland partitions correspond one-to-one. An optimization model construction module is used to construct a fractional structure optimization model based on the a effective agricultural datasets through an optimization model construction module. The fractional structure optimization model includes a basic objective function and a joint constraint system. Output weight coefficients and consumption weight coefficients are set. The basic objective function is weighted by the output weight coefficients and consumption weight coefficients to obtain a weighted objective function. The weighted objective function and the joint constraint system are combined to obtain a single-objective solution model. The global optimization solution module is used to optimize the single-objective solution model through the global optimization solution module to obtain the globally optimal configuration scheme and complete the agricultural resource allocation based on global optimization.

Images