Multi-target task optimization scheduling method based on agricultural machinery operation efficiency

Abstract

The invention relates to the technical field of agricultural machinery task scheduling, in particular to a multi-target task optimization scheduling method based on agricultural machinery operation efficiency. The method comprises the following steps of performing agricultural machine operation efficiency parameter measurement on each agricultural machine under different operation conditions to obtain an agricultural machine operation efficiency data set; performing agricultural machine capability modeling according to the agricultural machine operation efficiency data set to obtain an agricultural machine capability matrix; the method comprises the following steps: partitioning a farmland by using a remote sensing technology to obtain a farmland partition map; performing farmland operation task decomposition according to the farmland partition map, and generating a sub-task set to obtain the sub-task set; and according to the agricultural machine capability matrix, cooperative edge generation is carried out, cooperative edge cost weighting processing is carried out, and a weighted cooperative edge set is obtained. According to the method, the balance of the scheduling scheme among multiple targets is realized by constructing the agricultural machinery capability matrix, the cooperation relation graph and the multi-target optimization model, and the disadvantage that the existing method is insufficient in consideration of agricultural machinery cooperation operation is effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural machinery task scheduling, and in particular to a multi-objective task optimization scheduling method based on agricultural machinery operation efficiency. Background Art

[0002] With the improvement of agricultural mechanization, how to efficiently dispatch agricultural machinery resources has become an important issue in agricultural production. Traditional agricultural machinery dispatching methods mainly rely on experience and are difficult to adapt to the needs of modern agriculture for refined and intelligent management. In recent years, with the rapid development of technologies such as the Internet of Things, remote sensing technology, and artificial intelligence, multi-objective task optimization dispatching methods based on agricultural machinery operation efficiency have emerged. However, existing methods mainly focus on the operating efficiency of a single agricultural machine, and lack consideration of the collaborative operation of multiple agricultural machines, as shown below:

[0003] Lack of modeling of collaborative relationships among agricultural machinery: Many models treat each agricultural machine as an independent individual and do not consider the possible collaborative relationships between agricultural machinery, such as the coordination between combine harvesters and transport vehicles.

[0004] Collaboration costs are not considered: Even when agricultural machinery collaboration is taken into account, the additional costs brought about by collaboration are often ignored, such as communication costs and waiting time costs during collaborative operations.

[0005] Single optimization objective: Some methods only focus on a single objective, such as minimizing operation time, while ignoring other important objectives, such as agricultural machinery utilization, operation cost, etc. Summary of the invention

[0006] Based on this, it is necessary to provide a multi-objective task optimization scheduling method based on agricultural machinery operation efficiency to solve at least one of the above technical problems.

[0007] To achieve the above purpose, a multi-objective task optimization scheduling method based on agricultural machinery operation efficiency includes the following steps:

[0008] Step S1: measuring the operating efficiency parameters of each agricultural machine under different operating conditions to obtain an agricultural machine operating efficiency data set; modeling the agricultural machine capability based on the agricultural machine operating efficiency data set to obtain an agricultural machine capability matrix;

[0009] Step S2: using remote sensing technology to partition the farmland and obtain a farmland partition map; decomposing the farmland operation tasks according to the farmland partition map, and generating subtask sets to obtain subtask sets;

[0010] Step S3: generating collaborative edges according to the agricultural machinery capability matrix to obtain a collaborative edge set; performing collaborative edge cost weighting processing on the collaborative edge set according to the agricultural machinery capability matrix to obtain a weighted collaborative edge set; generating a collaborative relationship graph according to the weighted collaborative edge set to obtain a collaborative relationship graph;

[0011] Step S4: construct a multi-objective optimization model according to the collaboration relationship diagram to obtain a multi-objective optimization model; perform Pareto solution processing according to the multi-objective optimization model to obtain a resource scheduling solution set;

[0012] Step S5: using the simulation software and the resource scheduling solution set to run a simulation experiment and obtain simulation experiment result data; calculating the resource scheduling weight according to the simulation experiment result data, and generating a resource scheduling solution to obtain a resource scheduling solution;

[0013] Step S6: Import the resource scheduling plan into the agricultural machinery operation management system, dynamically process the scheduling plan, and obtain a dynamic scheduling trigger signal; perform dynamic scheduling execution using a multi-objective optimization model based on the dynamic scheduling trigger signal to obtain an agricultural machinery operation execution report.

[0014] The present invention can comprehensively grasp the operation capacity and performance characteristics of agricultural machinery by measuring the efficiency parameters of each agricultural machinery under different operation conditions and constructing an agricultural machinery capacity matrix, so as to provide an accurate data basis for subsequent scheduling optimization, thereby improving the allocation efficiency of agricultural machinery resources. By using remote sensing technology and agricultural activity demand data, farmland is partitioned and operation tasks are decomposed, and a subtask set containing detailed information is generated, so that large-scale farmland operation tasks can be refined into units that are easier to manage and schedule, providing the possibility for subsequent accurate scheduling. By constructing a collaborative relationship diagram including agricultural machinery, subtasks and the relationship between them, and cost-weighting the collaborative edges, the matching relationship between agricultural machinery and tasks and the collaborative relationship between agricultural machinery can be intuitively displayed, and the collaborative cost can be quantified, providing a visual analysis tool and an accurate cost reference for subsequent multi-objective optimization scheduling. According to the collaborative relationship diagram, a multi-objective optimization model is constructed, and a multi-objective evolutionary algorithm is used for solving. Multiple optimization objectives can be considered at the same time, such as minimizing operation time, cost and maximizing agricultural machinery utilization, and a set of Pareto optimal solutions that achieve a good balance between multiple objectives are found, providing decision makers with a variety of feasible scheduling schemes. By using simulation software to conduct simulation experiments on the Pareto optimal solution set, and calculating the comprehensive score of each solution based on the experimental result data and the evaluation index weight, a scheduling solution that is closer to the actual situation and has better comprehensive performance can be selected to improve the reliability and practicality of the solution. The finally selected scheduling solution is applied to actual operations, and real-time monitoring, anomaly detection and dynamic scheduling are performed. Various emergencies can be discovered and handled in a timely manner, and the scheduling solution can be dynamically adjusted according to changes in the actual situation to improve the adaptability and robustness of agricultural machinery operations, and finally realize the intelligent management of agricultural machinery operations, and improve the operation efficiency and economic benefits. Therefore, the present invention provides a multi-objective task optimization scheduling method based on agricultural machinery operation efficiency. By constructing an agricultural machinery capability matrix, the operation capability and collaborative relationship parameters of each agricultural machinery are comprehensively evaluated to provide a data basis for collaborative operation analysis. On this basis, through agricultural machinery-subtask matching and agricultural machinery-agricultural machinery collaborative relationship identification, a weighted collaborative edge set is generated, and the collaborative relationship and cost between multiple agricultural machines are incorporated into the scheduling model. Further, a collaborative relationship diagram is constructed to provide a visual analysis tool for multi-objective optimization scheduling, which is convenient for identifying and optimizing collaborative relationships. Finally, a multi-objective optimization model is adopted and multiple objective functions are defined, such as maximizing the utilization rate of agricultural machinery and minimizing operation time and cost, so as to finally achieve a balance between multiple objectives in the scheduling scheme, which effectively solves the problem that the existing methods do not give enough consideration to the collaborative operation of agricultural machinery.

[0015] Preferably, step S1 comprises the following steps:

[0016] Step S11: collecting basic information of each agricultural machine to obtain an agricultural machine basic information table; measuring agricultural machine operation efficiency parameters under different operation conditions according to the agricultural machine basic information table to obtain an agricultural machine operation efficiency data set;

[0017] Step S12: performing fuel consumption characteristic analysis based on the agricultural machinery operation efficiency data set and the agricultural machinery basic information table to obtain agricultural machinery fuel consumption characteristic data;

[0018] Step S13: constructing an agricultural machinery fuel consumption model according to the agricultural machinery fuel consumption characteristic data to obtain an agricultural machinery fuel consumption model;

[0019] Step S14: forecasting maintenance demand based on the agricultural machinery basic information table, the agricultural machinery operation efficiency data set, and the agricultural machinery fuel consumption model to obtain agricultural machinery maintenance demand planning data;

[0020] Step S15: calibrating the agricultural machinery cooperation relationship parameters for each agricultural machinery to obtain an agricultural machinery cooperation relationship parameter table;

[0021] Step S16: Generate an agricultural machinery capability matrix based on the agricultural machinery basic information table, the agricultural machinery operation efficiency data set, the agricultural machinery fuel consumption model, the agricultural machinery maintenance demand plan data, and the agricultural machinery cooperation relationship parameter table to obtain the agricultural machinery capability matrix.

[0022] The present invention can form an agricultural machinery operation efficiency data set by collecting basic information of each agricultural machinery and measuring the agricultural machinery operation efficiency parameters under different operating conditions, so as to provide basic data support for subsequent agricultural machinery capacity modeling and scheduling optimization. By analyzing the fuel consumption characteristics of the agricultural machinery operation efficiency data set and the agricultural machinery basic information table, the agricultural machinery fuel consumption characteristic data can be obtained, which provides a data basis for building an accurate agricultural machinery fuel consumption model, and then realizes the accurate evaluation of the agricultural machinery operation cost. The agricultural machinery fuel consumption model constructed according to the agricultural machinery fuel consumption characteristic data can accurately predict the fuel consumption of agricultural machinery under different operating conditions, and provide a scientific basis for optimizing the scheduling scheme and reducing the agricultural machinery operation cost. By combining the basic information of agricultural machinery, the operation efficiency data set and the fuel consumption model, the future maintenance needs of agricultural machinery can be predicted, and the agricultural machinery maintenance demand plan data can be generated, so as to realize the advance planning and preventive maintenance of agricultural machinery maintenance, reduce the failure rate of agricultural machinery, and improve the operation efficiency. The agricultural machinery cooperation relationship parameters of each agricultural machinery are calibrated, and the cooperation relationship between different agricultural machinery can be quantified, which provides data support for identifying the best cooperation combination and improving the efficiency of cooperative operation. By integrating basic agricultural machinery information, operation efficiency data sets, fuel consumption models, maintenance demand planning data, and agricultural machinery collaboration relationship parameters, an agricultural machinery capability matrix can be generated to provide a comprehensive and accurate data basis for agricultural machinery scheduling and management, thereby achieving optimal allocation and efficient utilization of agricultural machinery resources.

[0023] Preferably, step S2 comprises the following steps:

[0024] Step S21: using remote sensing technology to collect farmland information and obtain a farmland geographic information database;

[0025] Step S22: Acquire agricultural activity demand data; identify the operation type according to the agricultural activity demand data to obtain an operation type list;

[0026] Step S23: partitioning the farmland space according to the farmland geographic information database and the operation type list to obtain a farmland partition map;

[0027] Step S24: Decomposing farmland operation tasks according to the farmland zoning map and the farmland geographic information database to obtain a subtask spatial distribution map;

[0028] Step S25: assigning subtask attributes according to the subtask spatial distribution diagram to obtain a subtask attribute table; generating a subtask set according to the subtask attribute table to obtain a subtask set.

[0029] The present invention can quickly and accurately obtain high-precision farmland geographic information by using remote sensing technology to collect farmland information, and construct a farmland geographic information database containing information such as terrain, landform, soil, and crops, providing a detailed spatial data basis for the decomposition and scheduling of farmland operation tasks. The demand data of agricultural activities is obtained, and the types of agricultural machinery operations that need to be performed are identified therefrom to form a list of operation types, which can clarify specific operation goals and provide guidance for subsequent farmland spatial partitioning and task decomposition. According to the farmland geographic information database and the list of operation types, the farmland is spatially partitioned, and the areas that need to perform the same or similar types of operations can be divided together to form a farmland partition map, which provides more detailed spatial units for subsequent task decomposition. According to the farmland partition map and the farmland geographic information database, the farmland operation tasks are decomposed by comprehensively considering factors such as the workload, the operation time window, and the accessibility, and the subtask spatial distribution map is generated, which can decompose large-scale farmland operation tasks into subtasks that are easier to manage and schedule, thereby improving the scheduling efficiency. Assign attributes to each subtask, such as operation type, operation area, operation time window, etc., and generate a subtask attribute table, which can provide detailed information description for each subtask to facilitate subsequent agricultural machinery scheduling and management. Finally, integrate the information of all subtasks to generate a subtask set, providing structured data support for intelligent scheduling.

[0030] Preferably, step S24 includes the following steps:

[0031] Step S241: Estimating the workload according to the farmland zoning map to obtain a workload estimation table;

[0032] Step S242: determining the operation time window for the farmland partition map to obtain an operation time window table;

[0033] Step S243: performing obstacle identification on the farmland partition map to obtain an obstacle information layer;

[0034] Step S244: performing a traversability analysis based on the obstacle information layer and the farmland geographic information database to obtain a traversability analysis diagram;

[0035] Step S245: Perform subtask spatial clustering analysis based on the farmland zoning map, the workload estimation table, the operation time window table, and the accessibility analysis diagram to obtain a subtask spatial distribution map.

[0036] The present invention can clearly understand the workload size of each area by estimating the workload according to the farmland zoning map and generating a workload estimation table, providing a reference basis for subsequent agricultural machinery scheduling and resource allocation, and avoiding resource waste or shortage. By determining the operation time window of the farmland zoning map and generating an operation time window table, the optimal operation time range of each area can be clarified, which is convenient for subsequent agricultural machinery scheduling, improves operation efficiency, and ensures that the needs of crop growth are met. Obstacle identification is performed on the farmland zoning map, and an obstacle information layer is generated, which can clearly mark the location and type of obstacles in the farmland, provide a data basis for subsequent traversability analysis and path planning, avoid collision or detour during agricultural machinery operation, and improve operation safety and efficiency. According to the obstacle information layer and the farmland geographic information database, traversability analysis is performed, and a traversability analysis diagram is generated, which can intuitively display the traversable area of ​​agricultural machinery in the farmland, provide a reference for agricultural machinery path planning and scheduling, avoid agricultural machinery from entering inaccessible areas, and improve operation efficiency and safety. Based on the farmland zoning map, workload estimation table, operation time window table and accessibility analysis diagram, subtask spatial clustering analysis is performed and a subtask spatial distribution map is generated. Each area can be further divided into more detailed and reasonable subtasks, and the spatial position and boundaries of each subtask can be determined, providing more accurate data support for subsequent agricultural machinery scheduling and path planning, thereby improving scheduling efficiency and resource utilization.

[0037] Preferably, step S3 comprises the following steps:

[0038] Step S31: creating agricultural machinery nodes according to the agricultural machinery capability matrix to obtain an agricultural machinery node set; creating subtask nodes according to the subtask information in the subtask set to obtain a subtask node set;

[0039] Step S32: performing agricultural machinery-subtask matching according to the subtask node set and the agricultural machinery node set to obtain an agricultural machinery-subtask matching table;

[0040] Step S33: identifying the agricultural machinery-agricultural machinery cooperation relationship according to the agricultural machinery-subtask matching table and the agricultural machinery capability matrix, and obtaining the agricultural machinery-agricultural machinery cooperation relationship table;

[0041] Step S34: generating collaborative edges according to the agricultural machinery-subtask matching table and the agricultural machinery-agricultural machinery collaborative relationship table to obtain a collaborative edge set;

[0042] Step S35: performing collaborative edge cost weighting processing on the collaborative edge set according to the agricultural machinery-subtask matching table and the agricultural machinery-agricultural machinery collaborative relationship table to obtain a weighted collaborative edge set;

[0043] Step S36: Generate a collaboration relationship graph based on the agricultural machinery node set, the subtask node set, and the weighted collaboration edge set to obtain a collaboration relationship graph.

[0044] The present invention can abstract agricultural machinery and subtasks into nodes in graph theory by creating agricultural machinery node sets and subtask node sets respectively according to the agricultural machinery capability matrix and the subtask set, thus laying a foundation for the subsequent construction of a collaborative relationship graph. Agricultural machinery-subtask matching is performed according to the subtask node set and the agricultural machinery node set, and a matching table is generated, so that agricultural machinery that can complete specific subtasks can be screened out, and candidate agricultural machinery resources can be provided for subsequent collaborative relationship identification and scheduling optimization. Agricultural machinery-agricultural machinery collaborative relationships are identified according to the agricultural machinery-subtask matching table and the agricultural machinery capability matrix, and a collaborative relationship table is generated, so that it can be determined which agricultural machinery can collaboratively complete specific subtasks, fully tap the collaborative potential between agricultural machinery, and provide the possibility of improving work efficiency. A collaborative edge set is generated according to the agricultural machinery-subtask matching table and the agricultural machinery-agricultural machinery collaborative relationship table, so that the relationship between agricultural machinery and subtasks, and between agricultural machinery and agricultural machinery can be abstracted into edges in graph theory, providing a data basis for constructing a collaborative relationship graph. The collaborative edge set is weighted by the collaborative edge cost to obtain a weighted collaborative edge set, which can take into account factors such as time cost and fuel cost, so that the collaborative relationship diagram can better reflect the actual situation and provide a more accurate basis for subsequent scheduling optimization. The collaborative relationship diagram is generated based on the agricultural machinery node set, subtask node set and weighted collaborative edge set, which can intuitively display the agricultural machinery, subtasks and the relationship between them, providing a visual analysis tool for subsequent multi-objective optimization scheduling, and facilitating the identification and optimization of collaborative relationships.

[0045] Preferably, step S35 includes the following steps:

[0046] Step S351: Calculate the time cost according to the agricultural machine-subtask matching table and the agricultural machine-agricultural machine cooperation relationship table to obtain a time cost matrix;

[0047] Step S352: Calculate the fuel cost according to the agricultural machinery-subtask matching table and the agricultural machinery fuel consumption model in the agricultural machinery capability matrix to obtain a fuel cost matrix;

[0048] Step S353: Perform weighted sum analysis on the time cost matrix and the fuel cost matrix to obtain a collaboration cost matrix;

[0049] Step S354: assigning collaborative edge weights to the collaborative edge set according to the cost values ​​in the collaborative cost matrix to obtain a weighted collaborative edge set.

[0050] The present invention can quantify the time efficiency differences of different agricultural machinery-subtask combinations by calculating the time cost of each agricultural machinery to complete each subtask and generating a time cost matrix, thereby providing a basis for subsequent cost analysis and scheduling optimization. The fuel cost of each agricultural machinery to complete each subtask is calculated using the agricultural machinery fuel consumption model, and a fuel cost matrix is ​​generated, thereby quantifying the fuel consumption differences of different agricultural machinery-subtask combinations, thereby providing a basis for subsequent cost analysis and scheduling optimization. By performing weighted sum analysis on the time cost matrix and the fuel cost matrix, a collaborative cost matrix is ​​obtained, which can comprehensively consider the time cost and fuel cost, more comprehensively evaluate the cost-effectiveness of different agricultural machinery-subtask combinations, and provide a basis for subsequent collaborative edge weight assignment. By assigning collaborative edge weights to the collaborative edge set according to the cost values ​​in the collaborative cost matrix, the cost factors can be integrated into the collaborative relationship graph, so that the edge weights in the graph can reflect the cost differences in the real world, and provide a more accurate cost reference for subsequent multi-objective optimization scheduling, thereby better achieving cost control goals.

[0051] Preferably, step S4 comprises the following steps:

[0052] Step S41: define decision variables according to the collaboration diagram to obtain a decision variable set; construct an objective function according to the collaboration diagram to obtain an objective function expression; define constraint conditions according to the collaboration diagram to obtain a constraint condition set;

[0053] Step S42: constructing a multi-objective optimization model according to the decision variable set, the objective function expression and the constraint condition set to obtain a multi-objective optimization model;

[0054] Step S43: selecting a multi-objective evolutionary algorithm according to the multi-objective optimization model to obtain a selected multi-objective evolutionary algorithm; generating an initial population according to the selected multi-objective evolutionary algorithm and the multi-objective optimization model to obtain initial population data;

[0055] Step S44: Calculate the fitness function according to the objective function expression to obtain a fitness value set;

[0056] Step S45: using the selected multi-objective evolutionary algorithm to perform population evolution operations on the initial population data to obtain new population data; calculating the fitness value of the new population data to obtain the population fitness value;

[0057] Step S46: Performing Pareto dominance relationship judgment on the population fitness values ​​according to the fitness value set to obtain a Pareto dominance relationship matrix; extracting resource scheduling solutions according to the Pareto dominance relationship matrix to obtain a resource scheduling solution set.

[0058] The present invention can abstract the agricultural machinery scheduling problem into a mathematical optimization problem by defining decision variables, objective functions and constraints according to the collaborative relationship diagram, laying the foundation for the subsequent solution using the optimization algorithm. A multi-objective optimization model is constructed according to the decision variables, objective functions and constraints, and multiple optimization objectives and constraints in the actual problem can be described in mathematical language to form a solvable mathematical model, providing goals and directions for subsequent optimization solutions. According to the characteristics of the multi-objective optimization model, a suitable multi-objective evolutionary algorithm is selected, and initial population data is generated. The powerful global search ability of the evolutionary algorithm can be used to find solutions that meet the constraints and have better objective function values ​​in the solution space, providing an effective way to find the Pareto optimal solution set of the problem. The fitness value of each individual is calculated according to the objective function expression, and the optimization goal can be converted into a quantifiable indicator for evaluating the quality of each solution, providing a basis for the selection and evolution of the evolutionary algorithm. By using the selected multi-objective evolutionary algorithm to select, cross, mutate and perform other operations on the initial population, generate a new population, and calculate the fitness value of the new population, the population can be continuously iterated and optimized to evolve in the direction of the Pareto optimal solution set, and finally find a set of solutions that meet the constraints and have a better objective function value. By judging the Pareto dominance relationship of the population fitness value and extracting the Pareto optimal solution to form a resource scheduling solution set, a set of solutions that achieve a good balance between multiple objectives can be obtained, providing decision makers with a variety of feasible scheduling solutions so that they can choose according to actual conditions.

[0059] Preferably, step S5 comprises the following steps:

[0060] Step S51: constructing a farmland operation simulation environment using simulation software to obtain a virtual farmland operation environment;

[0061] Step S52: importing the resource scheduling solution set into the virtual farmland operation environment, setting the agricultural machinery operation parameters, and obtaining the agricultural machinery operation simulation parameters;

[0062] Step S53: performing simulation experiments according to the virtual farmland operation environment and the agricultural machinery operation simulation parameters to obtain simulation experiment result data;

[0063] Step S54: Calculate resource scheduling weights based on simulation experiment result data to obtain a scheduling scheme comprehensive scoring table;

[0064] Step S55: Generate a resource scheduling plan according to the scheduling plan comprehensive scoring table to obtain a resource scheduling plan.

[0065] The present invention uses simulation software to construct a virtual farmland operation environment, which can simulate the real farmland environment, agricultural machinery operation process and various influencing factors, provide a realistic simulation platform for evaluating and optimizing scheduling schemes, and avoid risks and costs in actual operations. The resource scheduling scheme set is imported into the virtual farmland operation environment and the corresponding agricultural machinery operation parameters are set. The theoretical scheduling scheme can be applied to the simulation environment for verification, and the feasibility and effectiveness of the scheme can be tested, providing a basis for subsequent simulation experiments. According to the virtual farmland operation environment and the agricultural machinery operation simulation parameters, the simulation experiment is run, the agricultural machinery operation process can be simulated, and various data such as operation time, fuel consumption, operation efficiency, etc. can be collected to provide a quantitative basis for the subsequent scheduling scheme evaluation. According to the simulation experiment result data, the resource scheduling weight is calculated, and a scheduling scheme comprehensive scoring table is generated. Multiple evaluation indicators can be comprehensively considered, and different scheduling schemes can be quantitatively evaluated to provide a decision-making basis for the selection of the final scheme. According to the scheduling scheme comprehensive scoring table, the scheme with the highest comprehensive score is selected as the final resource scheduling scheme, and the results of the simulation experiment can be fully utilized to select the optimal scheduling scheme, maximize the utilization rate of agricultural machinery, operation efficiency, and reduce operation costs.

[0066] Preferably, step S54 includes the following steps:

[0067] Step S541: Calculate the utilization rate of each agricultural machine under each scheduling scheme according to the simulation experiment result data to obtain an agricultural machine utilization rate statistics table;

[0068] Step S542: Calculate the operating efficiency of each subtask under each scheduling scheme according to the simulation experiment result data to obtain an operating efficiency statistics table;

[0069] Step S543: Calculate the operation cost of each subtask under each scheduling scheme according to the simulation experiment result data to obtain an operation cost statistics table;

[0070] Step S544: acquiring farmland operation demand data; determining the evaluation index weights according to the farmland operation demand data to obtain the evaluation index weight vector;

[0071] Step S545: Calculate the comprehensive score of the scheduling scheme according to the agricultural machinery utilization rate statistics table, the operation efficiency statistics table, the operation cost statistics table and the evaluation index weight vector to obtain a scheduling scheme comprehensive score table.

[0072] The present invention can quantitatively evaluate the utilization of agricultural machinery resources in each scheme by calculating the utilization rate of each agricultural machinery under each scheduling scheme and generating a statistical table of agricultural machinery utilization, so as to provide an important basis for subsequent comprehensive scoring, so as to select a scheme with higher resource utilization rate. By calculating the operating efficiency of each subtask under each scheduling scheme and generating a statistical table of operating efficiency, the operating efficiency of each scheme can be quantitatively evaluated, so as to provide an important basis for subsequent comprehensive scoring, so as to select a scheme with higher operating efficiency. By calculating the operating cost of each subtask under each scheduling scheme and generating a statistical table of operating cost, the economic benefit of each scheme can be quantitatively evaluated, so as to provide an important basis for subsequent comprehensive scoring, so as to select a scheme with lower operating cost. By obtaining farmland operation demand data and determining the evaluation index weight according to the demand data, the actual operation demand can be integrated into the evaluation system, so that the evaluation result is more in line with the actual situation. For example, if the operation time is urgent, the weight of the operation efficiency can be increased. The comprehensive score of each scheduling plan is calculated based on the agricultural machinery utilization statistics table, operation efficiency statistics table, operation cost statistics table and evaluation index weight vector, and a scheduling plan comprehensive score table is generated. It can comprehensively consider multiple evaluation indicators and make a more comprehensive and objective evaluation of different scheduling plans, providing a more reliable basis for the selection of the final plan.

[0073] Preferably, step S6 comprises the following steps:

[0074] Step S61: importing the resource scheduling plan into the agricultural machinery operation management system, generating agricultural machinery operation instructions, and obtaining agricultural machinery operation instructions; using the Internet of Things technology to monitor the agricultural machinery status and obtain the agricultural machinery status data stream;

[0075] Step S62: tracking the operation progress according to the agricultural machinery status data stream and the agricultural machinery operation instructions to obtain an operation progress report;

[0076] Step S63: performing abnormal event detection on the agricultural machinery status data stream to obtain abnormal event alarm data;

[0077] Step S64: acquiring real-time meteorological data; determining the dynamic scheduling triggering condition according to the abnormal event alarm data and the real-time meteorological data, and obtaining a dynamic scheduling triggering signal;

[0078] Step S65: Generate a dynamic scheduling solution using a multi-objective optimization model according to the dynamic scheduling trigger signal to obtain a dynamic scheduling solution;

[0079] Step S66: updating the agricultural machinery operation instruction according to the dynamic scheduling plan, and recording the operation data to obtain the operation record data; generating the agricultural machinery operation execution report according to the operation record data to obtain the agricultural machinery operation execution report.

[0080] The present invention can convert the scheduling scheme into executable instructions to guide the agricultural machinery to operate by importing the resource scheduling scheme into the agricultural machinery operation management system and generating agricultural machinery operation instructions; at the same time, the Internet of Things technology is used to monitor the state of agricultural machinery, so that the operation state of agricultural machinery can be grasped in real time, and data support can be provided for subsequent operation progress tracking and abnormal event detection. According to the agricultural machinery state data stream and the agricultural machinery operation instructions, the operation progress is tracked, and the operation progress report is generated, so that the completion status of each subtask can be understood in real time, and the potential progress delay risk can be discovered in time, providing a decision-making basis for managers. By detecting abnormal events on the agricultural machinery state data stream and generating abnormal event alarm data, agricultural machinery failures, operation abnormalities, etc. can be discovered in time, and relevant personnel can be notified in time to handle them, so as to avoid causing greater losses and improve the safety of agricultural machinery operations. By obtaining real-time meteorological data and combining the abnormal event alarm data to judge the trigger conditions of dynamic scheduling, it can be timely judged whether dynamic scheduling is needed according to the changes in actual conditions, such as sudden rainfall, agricultural machinery failure, etc., to provide a basis for the subsequent adjustment of the scheduling scheme. According to the dynamic scheduling trigger signal, a new scheduling plan is generated using a multi-objective optimization model. The original scheduling plan can be adjusted and optimized according to changes in the actual situation, improving the flexibility and adaptability of the scheduling plan and ensuring the smooth completion of agricultural machinery operations. According to the dynamic scheduling plan, the agricultural machinery operation instructions are updated and the operation data is recorded. The adjusted scheduling plan can be applied to the actual operation and the real operation data can be collected; the agricultural machinery operation execution report is generated based on the operation record data, and the agricultural machinery operation can be summarized and analyzed to provide data support for subsequent operation optimization and management.

[0081] The present invention can comprehensively grasp the operation capacity and performance characteristics of agricultural machinery by measuring the efficiency parameters of each agricultural machinery under different operation conditions and constructing an agricultural machinery capacity matrix, so as to provide an accurate data basis for subsequent scheduling optimization, thereby improving the allocation efficiency of agricultural machinery resources. By using remote sensing technology and agricultural activity demand data, farmland is partitioned and operation tasks are decomposed, and a subtask set containing detailed information is generated, so that large-scale farmland operation tasks can be refined into units that are easier to manage and schedule, providing the possibility for subsequent accurate scheduling. By constructing a collaborative relationship diagram including agricultural machinery, subtasks and the relationship between them, and cost-weighting the collaborative edges, the matching relationship between agricultural machinery and tasks and the collaborative relationship between agricultural machinery can be intuitively displayed, and the collaborative cost can be quantified, providing a visual analysis tool and an accurate cost reference for subsequent multi-objective optimization scheduling. According to the collaborative relationship diagram, a multi-objective optimization model is constructed, and a multi-objective evolutionary algorithm is used for solving. Multiple optimization objectives can be considered at the same time, such as minimizing operation time, cost and maximizing agricultural machinery utilization, and a set of Pareto optimal solutions that achieve a good balance between multiple objectives are found, providing decision makers with a variety of feasible scheduling schemes. By using simulation software to conduct simulation experiments on the Pareto optimal solution set, and calculating the comprehensive score of each solution based on the experimental result data and the evaluation index weight, a scheduling solution that is closer to the actual situation and has better comprehensive performance can be selected to improve the reliability and practicality of the solution. The finally selected scheduling solution is applied to actual operations, and real-time monitoring, anomaly detection and dynamic scheduling are performed. Various emergencies can be discovered and handled in a timely manner, and the scheduling solution can be dynamically adjusted according to changes in the actual situation to improve the adaptability and robustness of agricultural machinery operations, and finally realize the intelligent management of agricultural machinery operations, and improve the operation efficiency and economic benefits. Therefore, the present invention provides a multi-objective task optimization scheduling method based on agricultural machinery operation efficiency. By constructing an agricultural machinery capability matrix, the operation capability and collaborative relationship parameters of each agricultural machinery are comprehensively evaluated to provide a data basis for collaborative operation analysis. On this basis, through agricultural machinery-subtask matching and agricultural machinery-agricultural machinery collaborative relationship identification, a weighted collaborative edge set is generated, and the collaborative relationship and cost between multiple agricultural machines are incorporated into the scheduling model. Further, a collaborative relationship diagram is constructed to provide a visual analysis tool for multi-objective optimization scheduling, which is convenient for identifying and optimizing collaborative relationships. Finally, a multi-objective optimization model is adopted and multiple objective functions are defined, such as maximizing the utilization rate of agricultural machinery and minimizing operation time and cost, so as to finally achieve a balance between multiple objectives in the scheduling scheme, which effectively solves the problem that the existing methods do not give enough consideration to the collaborative operation of agricultural machinery. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 A schematic diagram of the steps of a multi-objective task optimization scheduling method based on agricultural machinery operation efficiency;

[0083] Figure 2 for Figure 1Detailed implementation steps of step S3 in FIG.

[0084] Figure 3 for Figure 1 Detailed implementation steps of step S4 in FIG.

[0085] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0086] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0087] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0088] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0089] To achieve this, please refer to Figures 1 to 3 , a multi-objective task optimization scheduling method based on agricultural machinery operation efficiency, comprising the following steps:

[0090] Step S1: measuring the operating efficiency parameters of each agricultural machine under different operating conditions to obtain an agricultural machine operating efficiency data set; modeling the agricultural machine capability based on the agricultural machine operating efficiency data set to obtain an agricultural machine capability matrix;

[0091] Step S2: using remote sensing technology to partition the farmland and obtain a farmland partition map; decomposing the farmland operation tasks according to the farmland partition map, and generating subtask sets to obtain subtask sets;

[0092] Step S3: generating collaborative edges according to the agricultural machinery capability matrix to obtain a collaborative edge set; performing collaborative edge cost weighting processing on the collaborative edge set according to the agricultural machinery capability matrix to obtain a weighted collaborative edge set; generating a collaborative relationship graph according to the weighted collaborative edge set to obtain a collaborative relationship graph;

[0093] Step S4: construct a multi-objective optimization model according to the collaboration relationship diagram to obtain a multi-objective optimization model; perform Pareto solution processing according to the multi-objective optimization model to obtain a resource scheduling solution set;

[0094] Step S5: using the simulation software and the resource scheduling solution set to run a simulation experiment and obtain simulation experiment result data; calculating the resource scheduling weight according to the simulation experiment result data, and generating a resource scheduling solution to obtain a resource scheduling solution;

[0095] Step S6: Import the resource scheduling plan into the agricultural machinery operation management system, dynamically process the scheduling plan, and obtain a dynamic scheduling trigger signal; perform dynamic scheduling execution using a multi-objective optimization model based on the dynamic scheduling trigger signal to obtain an agricultural machinery operation execution report.

[0096] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of the multi-objective task optimization scheduling method based on the agricultural machinery operation efficiency of the present invention. In this example, the multi-objective task optimization scheduling method based on the agricultural machinery operation efficiency includes the following steps:

[0097] Step S1: measuring the operating efficiency parameters of each agricultural machine under different operating conditions to obtain an agricultural machine operating efficiency data set; modeling the agricultural machine capability based on the agricultural machine operating efficiency data set to obtain an agricultural machine capability matrix;

[0098] In the embodiment of the present invention, by measuring the efficiency parameters of each agricultural machinery under different operating conditions and constructing an agricultural machinery capacity matrix, the operating capacity and performance characteristics of the agricultural machinery can be fully understood, providing an accurate data basis for subsequent scheduling optimization, thereby improving the allocation efficiency of agricultural machinery resources.

[0099] Step S2: using remote sensing technology to partition the farmland and obtain a farmland partition map; decomposing the farmland operation tasks according to the farmland partition map, and generating subtask sets to obtain subtask sets;

[0100] In an embodiment of the present invention, remote sensing technology and agricultural activity demand data are used to partition farmland and decompose work tasks, and a subtask set containing detailed information is generated. Large-scale farmland work tasks can be broken down into units that are easier to manage and schedule, providing the possibility for subsequent precise scheduling.

[0101] Step S3: generating collaborative edges according to the agricultural machinery capability matrix to obtain a collaborative edge set; performing collaborative edge cost weighting processing on the collaborative edge set according to the agricultural machinery capability matrix to obtain a weighted collaborative edge set; generating a collaborative relationship graph according to the weighted collaborative edge set to obtain a collaborative relationship graph;

[0102] In an embodiment of the present invention, by constructing a collaborative relationship diagram including agricultural machinery, subtasks and the relationships between them, and performing cost-weighting on the collaborative edges, the matching relationship between agricultural machinery and tasks and the collaborative relationship between agricultural machinery can be intuitively displayed, and the collaborative cost can be quantified, providing a visual analysis tool and accurate cost reference for subsequent multi-objective optimization scheduling.

[0103] Step S4: construct a multi-objective optimization model according to the collaboration relationship diagram to obtain a multi-objective optimization model; perform Pareto solution processing according to the multi-objective optimization model to obtain a resource scheduling solution set;

[0104] In an embodiment of the present invention, a multi-objective optimization model is constructed based on a collaborative relationship diagram, and a multi-objective evolutionary algorithm is used to solve it. Multiple optimization objectives can be considered at the same time, such as minimizing operation time and cost and maximizing agricultural machinery utilization, and a set of Pareto optimal solutions that achieve a good balance between multiple objectives can be found, providing decision makers with a variety of feasible scheduling solutions.

[0105] Step S5: using the simulation software and the resource scheduling solution set to run a simulation experiment and obtain simulation experiment result data; calculating the resource scheduling weight according to the simulation experiment result data, and generating a resource scheduling solution to obtain a resource scheduling solution;

[0106] In the embodiment of the present invention, simulation software is used to conduct simulation experiments on the Pareto optimal solution set, and the comprehensive score of each solution is calculated based on the experimental result data and the evaluation index weight. It is possible to select a scheduling solution that is closer to the actual situation and has better comprehensive performance, thereby improving the reliability and practicality of the solution.

[0107] Step S6: importing the resource scheduling plan into the agricultural machinery operation management system, dynamically processing the scheduling plan, and obtaining a dynamic scheduling trigger signal; performing dynamic scheduling execution using a multi-objective optimization model according to the dynamic scheduling trigger signal, and obtaining an agricultural machinery operation execution report;

[0108] In the embodiment of the present invention, the finally selected scheduling scheme is applied to the actual operation, and real-time monitoring, anomaly detection and dynamic scheduling are carried out, so as to timely discover and handle various emergencies, dynamically adjust the scheduling scheme according to the changes in the actual situation, improve the adaptability and robustness of agricultural machinery operations, and finally realize the intelligent management of agricultural machinery operations, thereby improving the operation efficiency and economic benefits.

[0109] Preferably, step S1 comprises the following steps:

[0110] Step S11: collecting basic information of each agricultural machine to obtain an agricultural machine basic information table; measuring agricultural machine operation efficiency parameters under different operation conditions according to the agricultural machine basic information table to obtain an agricultural machine operation efficiency data set;

[0111] Step S12: performing fuel consumption characteristic analysis based on the agricultural machinery operation efficiency data set and the agricultural machinery basic information table to obtain agricultural machinery fuel consumption characteristic data;

[0112] Step S13: constructing an agricultural machinery fuel consumption model according to the agricultural machinery fuel consumption characteristic data to obtain an agricultural machinery fuel consumption model;

[0113] Step S14: forecasting maintenance demand based on the agricultural machinery basic information table, the agricultural machinery operation efficiency data set, and the agricultural machinery fuel consumption model to obtain agricultural machinery maintenance demand planning data;

[0114] Step S15: calibrating the agricultural machinery cooperation relationship parameters for each agricultural machinery to obtain an agricultural machinery cooperation relationship parameter table;

[0115] Step S16: Generate an agricultural machinery capability matrix based on the agricultural machinery basic information table, the agricultural machinery operation efficiency data set, the agricultural machinery fuel consumption model, the agricultural machinery maintenance demand plan data, and the agricultural machinery cooperation relationship parameter table to obtain the agricultural machinery capability matrix.

[0116] In the embodiment of the present invention, first, basic information of each agricultural machine is collected using sensors and data acquisition equipment. The collected information includes, but is not limited to, agricultural machine ID, agricultural machine type (e.g., tractor, harvester, rice transplanter, etc.), manufacturer, model, production date, engine power, operating width, fuel tank capacity, current location and other information. The collected information is stored in a database to form an agricultural machine basic information table. Then, according to the agricultural machine basic information table, different operating conditions are selected to measure the agricultural machine operating efficiency parameters. For example, for tractors, different soil types, soil moisture, tillage depth and other conditions can be selected for tillage efficiency testing; for harvesters, different crop varieties, crop densities, crop heights and other conditions can be selected for harvesting efficiency testing. Under each operating condition, GPS positioning equipment and sensors are used to record the operating trajectory, operating area, operating time, fuel consumption and other data of the agricultural machine in real time. Finally, the collected data is processed and analyzed to obtain the operating efficiency parameters of different agricultural machines under different operating conditions, forming an agricultural machine operating efficiency data set. First, the engine power, operating width and other information of the agricultural machine are extracted from the agricultural machine basic information table. Then, combined with the agricultural machinery operation efficiency data set, the parameters such as the operation speed and operation load of agricultural machinery under different operation conditions are analyzed. Using the regression analysis method, the functional relationship between agricultural machinery fuel consumption and parameters such as operation speed, operation load, engine power, and operation width is established to obtain the characteristic data of agricultural machinery fuel consumption, such as: fuel consumption per unit area, fuel consumption per unit time, etc. According to the characteristic data of agricultural machinery fuel consumption, a suitable machine learning algorithm is selected to build an agricultural machinery fuel consumption model. For example, algorithms such as multivariate linear regression, support vector machine, and neural network can be used. The characteristic data of agricultural machinery fuel consumption is used as training data to train the selected algorithm to obtain a model that can predict the fuel consumption of agricultural machinery under different operating conditions. According to the information such as agricultural machinery type and production date in the agricultural machinery basic information table, combined with the data such as operation intensity and operation time in the agricultural machinery operation efficiency data set, and the fuel consumption predicted by the agricultural machinery fuel consumption model, the preset agricultural machinery maintenance rules and algorithms are used to predict the future maintenance needs of agricultural machinery, such as: engine maintenance, hydraulic oil replacement, blade replacement, etc. According to the prediction results, the agricultural machinery maintenance demand plan data is generated, including: expected maintenance time, maintenance items, required accessories and other information. Agricultural machinery is classified according to information such as its type and operating function. For example, tractors, seeders, and sprayers can be classified into one category, and combine harvesters can be classified into another category. For each type of agricultural machinery, its role and function in collaborative operations are analyzed, and the collaborative relationship parameters between agricultural machinery are defined. For example, the collaborative parameters between tractors and seeders can be defined as the degree of matching between the seeder's working efficiency and the tractor's driving speed, and the collaborative parameters between combine harvesters can be defined as the operating distance and operating time interval between combine harvesters. The collaborative relationship parameters of each agricultural machinery are calibrated through expert scoring, historical data analysis, and other methods to form a collaborative relationship parameter table for agricultural machinery.

[0117] The agricultural machinery capability matrix is ​​a multidimensional matrix used to store the comprehensive capability information of each agricultural machinery. The rows of the matrix represent different agricultural machinery, and the columns represent different capability indicators. First, the basic framework of the agricultural machinery capability matrix is ​​created based on the information such as the agricultural machinery ID and agricultural machinery type in the agricultural machinery basic information table. Then, the operating efficiency parameters under different operating conditions in the agricultural machinery operating efficiency data set obtained in step S11 are filled into the agricultural machinery and operating type indicators corresponding to the agricultural machinery capability matrix. Next, the agricultural machinery fuel consumption model obtained in step S13 is integrated into the agricultural machinery capability matrix as the agricultural machinery fuel consumption capability indicator. At the same time, the agricultural machinery maintenance demand planning data obtained in step S14 is converted into the agricultural machinery available time indicator and added to the agricultural machinery capability matrix. Finally, the collaboration parameters in the agricultural machinery collaboration relationship parameter table obtained in step S15 are also added to the agricultural machinery capability matrix as the agricultural machinery collaboration capability indicator. The agricultural machinery capability matrix finally obtained contains the basic information, operating efficiency, fuel consumption, maintenance requirements, collaboration relationship and other capability indicators of each agricultural machinery, providing a comprehensive data basis for subsequent agricultural machinery scheduling.

[0118] Preferably, step S2 comprises the following steps:

[0119] Step S21: using remote sensing technology to collect farmland information and obtain a farmland geographic information database;

[0120] Step S22: Acquire agricultural activity demand data; identify the operation type according to the agricultural activity demand data to obtain an operation type list;

[0121] Step S23: partitioning the farmland space according to the farmland geographic information database and the operation type list to obtain a farmland partition map;

[0122] Step S24: Decomposing farmland operation tasks according to the farmland zoning map and the farmland geographic information database to obtain a subtask spatial distribution map;

[0123] Step S25: assigning subtask attributes according to the subtask spatial distribution diagram to obtain a subtask attribute table; generating a subtask set according to the subtask attribute table to obtain a subtask set.

[0124] In an embodiment of the present invention, a multispectral camera, a laser radar and other sensors are used to collect data on the target farmland by using an unmanned aerial vehicle. High-resolution farmland remote sensing image data and point cloud data are obtained. Professional remote sensing image processing software is used to perform preprocessing operations such as geometric correction, radiation correction, and orthorectification on the acquired remote sensing image. Digital elevation model (DEM) data is generated using point cloud data. The preprocessed remote sensing image data, DEM data and farmland plot boundary information are imported into a geographic information system (GIS) platform to generate a farmland geographic information database containing information such as farmland boundaries, topography, landforms, soil types, and crop types. Agricultural activity demand data is obtained through an agricultural Internet of Things platform, an agricultural management system or manual survey. Agricultural activity demand data includes but is not limited to: planting plans, fertilization plans, pest and disease control plans, etc. According to a preset rule base, the agricultural activity demand data is analyzed to identify the type of agricultural machinery operation that needs to be performed. For example, according to information such as crop type, sowing time, and sowing depth in the planting plan, the type of agricultural machinery operation that needs to be performed is identified as "sowing". All identified agricultural machinery operation types are summarized to form an operation type list. In the GIS platform, the farmland data in the farmland geographic information database are spatially partitioned according to the list of operation types. For example, the plots that need to be "sowed" are divided into one area, and the plots that need to be "sprayed pesticides" are divided into another area. When performing spatial partitioning, factors such as the connectivity of the plots, the urgency of the operation, and the attribute information of the farmland need to be considered to ensure that the final generated farmland partition map can meet the actual needs of agricultural machinery operations. First, according to the farmland partition map, the operation area of ​​each area is calculated. Then, according to the agricultural machinery operation efficiency parameters, the operation time of each area is estimated. Then, the spatial analysis function of the GIS platform is used to identify obstacles in the farmland partition map, such as buildings, rivers, trees, etc., and an obstacle information layer is generated. According to the obstacle information layer and the road information in the farmland geographic information database, the passability of agricultural machinery in the farmland is analyzed, and a passability analysis map is generated. Finally, considering the operation area, operation time, passability and other factors, the cluster analysis algorithm is used to further divide each area into several subtasks, and a subtask spatial distribution map is generated. According to the spatial distribution map of subtasks, combined with the farmland geographic information database and agricultural activity demand data, specific attribute information is assigned to each subtask, such as operation type, operation area, operation time window, soil type, crop type, fertilizer application amount, pesticide type, etc. The attribute information of each subtask is stored in the subtask attribute table. Finally, according to the subtask attribute table, the information of all subtasks is integrated to generate a subtask set. The subtask set contains detailed information of each subtask, providing basic data for subsequent agricultural machinery scheduling.

[0125] Preferably, step S24 includes the following steps:

[0126] Step S241: Estimating the workload according to the farmland zoning map to obtain a workload estimation table;

[0127] Step S242: determining the operation time window for the farmland partition map to obtain an operation time window table;

[0128] Step S243: performing obstacle identification on the farmland partition map to obtain an obstacle information layer;

[0129] Step S244: performing a traversability analysis based on the obstacle information layer and the farmland geographic information database to obtain a traversability analysis diagram;

[0130] Step S245: Perform subtask spatial clustering analysis based on the farmland zoning map, the workload estimation table, the operation time window table, and the accessibility analysis diagram to obtain a subtask spatial distribution map.

[0131] In an embodiment of the present invention, in a GIS platform, an area calculation tool is called to calculate the area of ​​each area in a farmland zoning map. According to the operation parameters of each operation type preset in the agricultural activity demand data, such as sowing amount, fertilizer amount, spraying amount, etc., combined with the area of ​​each area, the operation amount of each area is calculated. For example, assuming that the area of ​​a certain area is 10 hectares, the operation type to be performed is "sowing", and the preset sowing amount is 20 kilograms per hectare, then the sowing operation amount of the area is 200 kilograms. The operation amount of each area is summarized into an operation amount estimation table, which includes information such as area ID, operation type, and operation amount. According to the optimal operation time range of each operation type in the agricultural activity demand data, meteorological data, and soil moisture information in the farmland geographic information database, the operation time window of each area is determined. For example, assuming that the operation type to be performed in a certain area is "sowing", the optimal sowing time range is from April 1 to April 10, and the meteorological conditions and soil moisture conditions in this time period meet the sowing requirements, then the operation time window of the area is from April 1 to April 10. The operation time window of each area is summarized into the operation time window table, which includes information such as area ID, operation type, operation start time, and operation end time. Use existing obstacle data, such as building outlines, road networks, water systems, etc., or use deep learning models to identify targets on high-resolution remote sensing images to identify obstacles in the farmland zoning map, such as buildings, trees, electric poles, rivers, etc. The identified obstacles are vectorized to generate an obstacle information layer, which includes information such as the type, location, shape, and size of the obstacles. In the GIS platform, spatial analysis tools are used to analyze the accessibility of agricultural machinery in the farmland based on the obstacle information layer and the road information in the farmland geographic information database. For example, the cost distance analysis method can be used to calculate the shortest path and travel cost for agricultural machinery from the ridge to any point in each area, and the accessible area and inaccessible area are divided according to the travel cost to generate a accessibility analysis diagram. In the GIS platform, the farmland zoning map, workload estimation table, operation time window table, and accessibility analysis map are comprehensively considered, and each area is further divided into several subtasks using the spatial clustering analysis algorithm. For example, an area can be divided into multiple subtasks with similar areas, regular shapes, and easy agricultural machinery operations based on workload and accessibility. When conducting spatial clustering analysis, factors such as the operating efficiency, operating cost, and agricultural machinery scheduling of the subtasks need to be considered to ensure that the final generated subtask spatial distribution map can meet the actual needs of agricultural machinery operations.

[0132] Preferably, step S3 comprises the following steps:

[0133] Step S31: creating agricultural machinery nodes according to the agricultural machinery capability matrix to obtain an agricultural machinery node set; creating subtask nodes according to the subtask information in the subtask set to obtain a subtask node set;

[0134] Step S32: performing agricultural machinery-subtask matching according to the subtask node set and the agricultural machinery node set to obtain an agricultural machinery-subtask matching table;

[0135] Step S33: identifying the agricultural machinery-agricultural machinery cooperation relationship according to the agricultural machinery-subtask matching table and the agricultural machinery capability matrix, and obtaining the agricultural machinery-agricultural machinery cooperation relationship table;

[0136] Step S34: generating collaborative edges according to the agricultural machinery-subtask matching table and the agricultural machinery-agricultural machinery collaborative relationship table to obtain a collaborative edge set;

[0137] Step S35: performing collaborative edge cost weighting processing on the collaborative edge set according to the agricultural machinery-subtask matching table and the agricultural machinery-agricultural machinery collaborative relationship table to obtain a weighted collaborative edge set;

[0138] Step S36: Generate a collaboration relationship graph based on the agricultural machinery node set, the subtask node set, and the weighted collaboration edge set to obtain a collaboration relationship graph.

[0139] As an example of the present invention, refer to Figure 2 As shown, in this example, step S3 includes:

[0140] Step S31: creating agricultural machinery nodes according to the agricultural machinery capability matrix to obtain an agricultural machinery node set; creating subtask nodes according to the subtask information in the subtask set to obtain a subtask node set;

[0141] In the embodiment of the present invention, the agricultural machinery capacity matrix is ​​traversed, and a corresponding agricultural machinery node is created for each agricultural machinery. Each agricultural machinery node contains information such as agricultural machinery ID, agricultural machinery type, operation efficiency, fuel consumption model, maintenance demand planning data, current location, etc. All created agricultural machinery nodes are stored in the agricultural machinery node set. The subtask set is traversed, and a corresponding subtask node is created for each subtask. Each subtask node contains information such as subtask ID, operation type, operation area, operation time window, geographical location, required agricultural machinery type, etc. All created subtask nodes are stored in the subtask node set.

[0142] Step S32: performing agricultural machinery-subtask matching according to the subtask node set and the agricultural machinery node set to obtain an agricultural machinery-subtask matching table;

[0143] In the embodiment of the present invention, the subtask node set and the agricultural machinery node set are matched according to the pre-set rules. For example, the agricultural machinery type required in the subtask node and the agricultural machinery type in the agricultural machinery node can be matched to screen out the agricultural machinery that can complete the operation of this type. At the same time, the matching results are sorted considering the operating efficiency, geographical location, time window and other factors of the agricultural machinery, and the agricultural machinery with high efficiency, short distance and time matching is preferentially selected. The matching results are stored in the agricultural machinery-subtask matching table, which contains information such as subtask ID, agricultural machinery ID, matching degree and so on.

[0144] Step S33: identifying the agricultural machinery-agricultural machinery cooperation relationship according to the agricultural machinery-subtask matching table and the agricultural machinery capability matrix, and obtaining the agricultural machinery-agricultural machinery cooperation relationship table;

[0145] In the embodiment of the present invention, the agricultural machinery-subtask matching table is traversed, and the agricultural machinery combination that can collaboratively complete the subtask is identified according to the agricultural machinery cooperation relationship parameters in the agricultural machinery capability matrix. For example, assuming that subtask A requires two operations, "sowing" and "fertilizing", the agricultural machinery-subtask matching table shows that agricultural machinery 1 can perform "sowing" and agricultural machinery 2 can perform "fertilization", and the agricultural machinery capability matrix shows that agricultural machinery 1 and agricultural machinery 2 can work collaboratively, then agricultural machinery 1 and agricultural machinery 2 are identified as agricultural machinery combinations that can collaboratively complete subtask A. All identified agricultural machinery-agricultural machinery cooperation relationships are stored in the agricultural machinery-agricultural machinery cooperation relationship table, which contains information such as subtask ID, cooperation agricultural machinery ID1, cooperation agricultural machinery ID2, cooperation relationship parameters, etc.

[0146] Step S34: generating collaborative edges according to the agricultural machinery-subtask matching table and the agricultural machinery-agricultural machinery collaborative relationship table to obtain a collaborative edge set;

[0147] In the embodiment of the present invention, a collaborative edge is created between the agricultural machinery node and the subtask node according to the agricultural machinery-subtask matching table and the agricultural machinery-agricultural machinery collaborative relationship table. For example, if agricultural machinery 1 matches subtask A, a collaborative edge is created between agricultural machinery node 1 and subtask node A. If agricultural machinery 1 and agricultural machinery 2 can collaboratively complete subtask A, a collaborative edge is created between agricultural machinery node 1 and agricultural machinery node 2. All created collaborative edges are stored in the collaborative edge set, and each edge contains information such as the starting node ID and the target node ID.

[0148] Step S35: performing collaborative edge cost weighting processing on the collaborative edge set according to the agricultural machinery-subtask matching table and the agricultural machinery-agricultural machinery collaborative relationship table to obtain a weighted collaborative edge set;

[0149] In the embodiment of the present invention, the cost of each collaborative edge, such as time cost, fuel cost, etc., is calculated based on the agricultural machinery-subtask matching table, the agricultural machinery-agricultural machinery collaborative relationship table, and the fuel consumption model in the agricultural machinery capability matrix. For example, the time cost can be calculated based on the agricultural machinery operation efficiency and the subtask workload, and the fuel cost can be calculated based on the agricultural machinery fuel consumption model and the operation distance. The calculated cost is used as a weight and assigned to each collaborative edge to obtain a weighted collaborative edge set.

[0150] Step S36: generating a collaboration relationship graph according to the agricultural machinery node set, the subtask node set and the weighted collaboration edge set to obtain a collaboration relationship graph;

[0151] In the embodiment of the present invention, a graph theory algorithm is used to construct a collaboration relationship graph based on the agricultural machinery node set, the subtask node set and the weighted collaboration edge set. In the collaboration relationship graph, the agricultural machinery nodes and the subtask nodes are represented by different graphics, and the collaboration edges are represented by weighted line segments. The collaboration relationship graph intuitively shows the matching relationship between agricultural machinery and subtasks and the collaboration relationship between agricultural machinery, providing a visual analysis tool for subsequent multi-objective optimization scheduling.

[0152] Preferably, step S35 includes the following steps:

[0153] Step S351: Calculate the time cost according to the agricultural machine-subtask matching table and the agricultural machine-agricultural machine cooperation relationship table to obtain a time cost matrix;

[0154] Step S352: Calculate the fuel cost according to the agricultural machinery-subtask matching table and the agricultural machinery fuel consumption model in the agricultural machinery capability matrix to obtain a fuel cost matrix;

[0155] Step S353: Perform weighted sum analysis on the time cost matrix and the fuel cost matrix to obtain a collaboration cost matrix;

[0156] Step S354: assigning collaborative edge weights to the collaborative edge set according to the cost values ​​in the collaborative cost matrix to obtain a weighted collaborative edge set.

[0157] In an embodiment of the present invention, a time cost matrix is ​​created with agricultural machinery and subtasks as row and column indexes. The agricultural machinery-subtask matching table is traversed, and the operating efficiency parameters of the agricultural machinery are obtained from the agricultural machinery capacity matrix according to the matched agricultural machinery and subtask information. The time required to complete the subtask is calculated in combination with the operating area of ​​the subtask, and the time is filled in the corresponding position of the time cost matrix. For example, if agricultural machinery 1 matches subtask A, according to the agricultural machinery capacity matrix, the operating efficiency of agricultural machinery 1 to complete the corresponding operation type of subtask A is 2 mu / hour, and the operating area of ​​subtask A is 5 mu. The time cost of agricultural machinery 1 to complete subtask A is 2.5 hours, and 2.5 is filled in the cells corresponding to agricultural machinery 1 and subtask A in the time cost matrix. For agricultural machinery-subtask combinations that cannot complete this type of operation, a maximum value is filled in the corresponding cell to indicate that it is not feasible. Create a fuel cost matrix with agricultural machinery and subtasks as row and column indexes. Traverse the agricultural machinery-subtask matching table, obtain the fuel consumption model of the agricultural machinery from the agricultural machinery capability matrix according to the matched agricultural machinery and subtask information, calculate the estimated fuel consumption of the agricultural machinery to complete the subtask in combination with the geographical location information of the subtask, and calculate the fuel cost according to the current oil price, and fill it into the corresponding position of the fuel cost matrix. For example, if agricultural machinery 1 matches subtask A, according to the fuel consumption model in the agricultural machinery capability matrix, the estimated fuel consumption of agricultural machinery 1 to complete subtask A is 5 liters, and the current oil price is 8 yuan / liter, then the fuel cost of agricultural machinery 1 to complete subtask A is 40 yuan, and 40 is filled into the corresponding cells of agricultural machinery 1 and subtask A in the fuel cost matrix. For agricultural machinery-subtask combinations that cannot complete this type of operation, fill in a maximum value in the corresponding cell to indicate that it is not feasible. Set the weight coefficients of time cost and fuel cost, for example, 0.6 and 0.4 respectively, indicating that more emphasis is placed on time cost. Multiply the elements of the corresponding positions of the time cost matrix and the fuel cost matrix by the corresponding weight coefficients and add them together to obtain a new matrix, namely the collaboration cost matrix. The value of each element in the collaboration cost matrix represents the comprehensive cost of the corresponding agricultural machinery-subtask combination. Traverse each edge in the collaboration edge set, and obtain the corresponding cost value from the collaboration cost matrix according to the starting node and target node of the edge. If the edge connects an agricultural machinery node and a subtask node, the obtained cost value is directly assigned to the edge as a weight; if the edge connects two agricultural machinery nodes, the cost value is added to the preset collaboration cost coefficient and assigned to the edge as a weight to reflect the additional cost brought by the collaboration. Finally, a weighted collaboration edge set is obtained, and each edge contains specific cost information.

[0158] Preferably, step S4 comprises the following steps:

[0159] Step S41: define decision variables according to the collaboration diagram to obtain a decision variable set; construct an objective function according to the collaboration diagram to obtain an objective function expression; define constraint conditions according to the collaboration diagram to obtain a constraint condition set;

[0160] Step S42: constructing a multi-objective optimization model according to the decision variable set, the objective function expression and the constraint condition set to obtain a multi-objective optimization model;

[0161] Step S43: selecting a multi-objective evolutionary algorithm according to the multi-objective optimization model to obtain a selected multi-objective evolutionary algorithm; generating an initial population according to the selected multi-objective evolutionary algorithm and the multi-objective optimization model to obtain initial population data;

[0162] Step S44: Calculate the fitness function according to the objective function expression to obtain a fitness value set;

[0163] Step S45: using the selected multi-objective evolutionary algorithm to perform population evolution operations on the initial population data to obtain new population data; calculating the fitness value of the new population data to obtain the population fitness value;

[0164] Step S46: Performing Pareto dominance relationship judgment on the population fitness values ​​according to the fitness value set to obtain a Pareto dominance relationship matrix; extracting resource scheduling solutions according to the Pareto dominance relationship matrix to obtain a resource scheduling solution set.

[0165] As an example of the present invention, refer to Figure 3 As shown, in this example, step S4 includes:

[0166] Step S41: define decision variables according to the collaboration diagram to obtain a decision variable set; construct an objective function according to the collaboration diagram to obtain an objective function expression; define constraint conditions according to the collaboration diagram to obtain a constraint condition set;

[0167] In an embodiment of the present invention, each edge in the collaborative relationship graph is regarded as a decision variable, and a binary variable is used to indicate whether the edge is selected. For example, a variable value of 1 indicates that the edge is selected, and 0 indicates that it is not selected. The decision variables of all edges constitute a decision variable set. According to actual needs, multiple objective functions are defined. For example, minimizing the total operation time, minimizing the total fuel cost, and maximizing the utilization rate of agricultural machinery are used as goals, and corresponding objective function expressions are constructed. The objective function expression should contain decision variables. For example, the total operation time can be expressed as the sum of the time costs of all selected edges. According to actual conditions, constraints are defined. For example, each subtask can only be assigned to one agricultural machine to complete, and each agricultural machine can only complete one subtask at the same time. Constraints are expressed in mathematical expressions. For example, for each subtask node, the number of selected edges connected to it cannot exceed 1.

[0168] Step S42: constructing a multi-objective optimization model according to the decision variable set, the objective function expression and the constraint condition set to obtain a multi-objective optimization model;

[0169] In the embodiment of the present invention, a multi-objective optimization model is constructed according to the decision variable set, objective function expression and constraint condition set defined in step S41. The model can be expressed in the form of mathematical programming, for example, the objective function expression is used as the objective function, the constraint condition is used as the constraint condition, and the value range of the decision variable is limited to 0 or 1.

[0170] Step S43: selecting a multi-objective evolutionary algorithm according to the multi-objective optimization model to obtain a selected multi-objective evolutionary algorithm; generating an initial population according to the selected multi-objective evolutionary algorithm and the multi-objective optimization model to obtain initial population data;

[0171] In the embodiment of the present invention, according to the characteristics of the constructed multi-objective optimization model, a suitable multi-objective evolutionary algorithm is selected, such as the NSGA-II algorithm, the MOEA / D algorithm, etc. According to the requirements of the selected algorithm and the scale of the model, the algorithm parameters are set, such as the population size, the number of iterations, the crossover probability, the mutation probability, etc. The initial population data satisfying the constraints is generated by random generation, and each individual represents a resource scheduling scheme, which is represented by a set of decision variable values.

[0172] Step S44: Calculate the fitness function according to the objective function expression to obtain a fitness value set;

[0173] In the embodiment of the present invention, according to the objective function expression defined in step S41, the function value of each individual on all objective functions is calculated to obtain a fitness value vector. For example, if the total operation time of the resource scheduling scheme represented by an individual is 10 hours, the total fuel cost is 800 yuan, and the agricultural machinery utilization rate is 85%, then the fitness value vector of the individual is (10,800,0.85).

[0174] Step S45: using the selected multi-objective evolutionary algorithm to perform population evolution operations on the initial population data to obtain new population data; calculating the fitness value of the new population data to obtain the population fitness value;

[0175] In the embodiment of the present invention, the selected multi-objective evolutionary algorithm selected in step S43 is used to perform operations such as selection, crossover, and mutation on the initial population data to generate new population data. For example, a roulette wheel selection method can be used to select excellent individuals for crossover and mutation to generate new individuals. For the newly generated population data, the fitness value is calculated according to the method in step S44 to obtain the population fitness value.

[0176] Step S46: Performing Pareto dominance relationship judgment on the fitness values ​​of the population according to the fitness value set to obtain a Pareto dominance relationship matrix; extracting resource scheduling solutions according to the Pareto dominance relationship matrix to obtain a resource scheduling solution set;

[0177] In the embodiment of the present invention, the fitness value vectors of all individuals are compared pairwise to determine the Pareto dominance relationship and construct a Pareto dominance relationship matrix. For example, if the fitness value vector of individual A is (10,800,0.85) and the fitness value vector of individual B is (12,750,0.80), then individual A is better than individual B in terms of total operation time and agricultural machinery utilization, and individual B is better than individual A in terms of total fuel cost. Therefore, there is no Pareto dominance relationship between individual A and individual B. According to the Pareto dominance relationship matrix, individuals that are not dominated by other individuals, that is, Pareto optimal solutions, are extracted to form a resource scheduling solution set.

[0178] Preferably, step S5 comprises the following steps:

[0179] Step S51: constructing a farmland operation simulation environment using simulation software to obtain a virtual farmland operation environment;

[0180] Step S52: importing the resource scheduling solution set into the virtual farmland operation environment, setting the agricultural machinery operation parameters, and obtaining the agricultural machinery operation simulation parameters;

[0181] Step S53: performing simulation experiments according to the virtual farmland operation environment and the agricultural machinery operation simulation parameters to obtain simulation experiment result data;

[0182] Step S54: Calculate resource scheduling weights based on simulation experiment result data to obtain a scheduling scheme comprehensive scoring table;

[0183] Step S55: Generate a resource scheduling plan according to the scheduling plan comprehensive scoring table to obtain a resource scheduling plan.

[0184] In an embodiment of the present invention, a virtual farmland operation environment is constructed by using agricultural machinery simulation software, such as Unity3D, Unreal Engine, etc., combined with information such as terrain, landform, obstacles, etc. in a farmland geographic information database. In the virtual environment, a virtual farmland is created according to the size and shape of the actual farmland, and parameters such as the terrain undulation, soil type, and crop type of the virtual farmland are set according to the data in the farmland geographic information database. At the same time, virtual obstacles such as houses, trees, rivers, etc., as well as information such as the operation path of agricultural machinery are added to the virtual farmland environment to simulate the real farmland operation scene. The resource scheduling scheme set obtained in step S4 is imported into the virtual farmland operation environment. According to the allocation of agricultural machinery and subtasks in each resource scheduling scheme, the virtual agricultural machinery is placed in the corresponding position of the virtual farmland environment, and the subtask information is associated with the virtual agricultural machinery. According to the agricultural machinery capability matrix and the subtask attribute table, the operation parameters of the virtual agricultural machinery are set, such as the operation speed, operation width, fuel consumption rate, etc., as well as the operation area and operation time window of the subtask. In the virtual farmland operation environment, according to the agricultural machinery operation parameters set in step S52, the simulation experiment is run. During the simulation experiment, the operation trajectory, operation time, fuel consumption and other data of each virtual agricultural machine under each resource scheduling scheme, as well as the completion time, operation quality and other data of each subtask are recorded. All simulation experiment result data are stored in the database for subsequent analysis and evaluation. According to the simulation experiment result data obtained in step S53, the comprehensive score of each resource scheduling scheme is calculated. For example, each scheme can be evaluated according to indicators such as agricultural machinery utilization rate, operation efficiency, and operation cost, and a corresponding weight can be assigned to obtain a comprehensive score of each scheme by weighted calculation. The comprehensive scores of all schemes are summarized in the scheduling scheme comprehensive score table, which contains information such as scheme ID, agricultural machinery utilization rate, operation efficiency, operation cost, and comprehensive score. According to the scheduling scheme comprehensive score table, the scheme with the highest comprehensive score is selected as the final resource scheduling scheme. For example, according to a preset threshold, schemes with a comprehensive score higher than the threshold can be screened out, and the scheme with the highest comprehensive score can be selected as the final scheme. If there are multiple schemes with the same comprehensive score and all are the highest, the scheme that meets other constraints, such as the shortest operation time and the lowest cost, can be selected as the final scheme.

[0185] Preferably, step S54 includes the following steps:

[0186] Step S541: Calculate the utilization rate of each agricultural machine under each scheduling scheme according to the simulation experiment result data to obtain an agricultural machine utilization rate statistics table;

[0187] Step S542: Calculate the operating efficiency of each subtask under each scheduling scheme according to the simulation experiment result data to obtain an operating efficiency statistics table;

[0188] Step S543: Calculate the operation cost of each subtask under each scheduling scheme according to the simulation experiment result data to obtain an operation cost statistics table;

[0189] Step S544: acquiring farmland operation demand data; determining the evaluation index weights according to the farmland operation demand data to obtain the evaluation index weight vector;

[0190] Step S545: Calculate the comprehensive score of the scheduling scheme according to the agricultural machinery utilization rate statistics table, the operation efficiency statistics table, the operation cost statistics table and the evaluation index weight vector to obtain a scheduling scheme comprehensive score table.

[0191] In an embodiment of the present invention, according to the simulation experiment result data saved in step S53, the actual operation time and total available time of each agricultural machine under each resource scheduling scheme are counted. The actual operation time of each agricultural machine is divided by the total available time to obtain the utilization rate of the agricultural machine under the scheme. The utilization rate data of all agricultural machines are summarized into an agricultural machine utilization rate statistical table, which contains information such as scheme ID, agricultural machine ID, and utilization rate. According to the simulation experiment result data saved in step S53, the actual completion time and operation area of ​​each subtask under each resource scheduling scheme are counted. The operation area of ​​each subtask is divided by the actual completion time to obtain the operation efficiency of the subtask under the scheme. The operation efficiency data of all subtasks are summarized into an operation efficiency statistical table, which contains information such as scheme ID, subtask ID, and operation efficiency. According to the simulation experiment result data saved in step S53, combined with preset cost parameters, such as labor cost per unit time, unit fuel consumption cost, etc., the operation cost of each subtask under each resource scheduling scheme is calculated. For example, the actual completion time of each subtask is multiplied by the labor cost per unit time, and the actual fuel consumption is multiplied by the fuel consumption cost per unit to obtain the operation cost of the subtask. The operation cost data of all subtasks are summarized into the operation cost statistics table, which contains information such as scheme ID, subtask ID, and operation cost. The farmland operation demand data is obtained through the agricultural Internet of Things platform, agricultural management system, or expert experience, such as the urgency of the operation, the operation quality requirements, and the cost control target. According to the obtained farmland operation demand data, the weights of evaluation indicators such as agricultural machinery utilization rate, operation efficiency, and operation cost are determined. For example, if the urgency of the operation is high, a higher weight is given to the operation efficiency. The weights of all evaluation indicators constitute the evaluation indicator weight vector. For example, (0.4, 0.5, 0.1) means that the weights of agricultural machinery utilization rate, operation efficiency, and operation cost are 0.4, 0.5, and 0.1 respectively. According to the agricultural machinery utilization statistics table, operation efficiency statistics table, operation cost statistics table obtained in steps S541, S542, and S543, and the evaluation index weight vector obtained in step S544, the comprehensive score of each resource scheduling scheme is calculated. For example, the average utilization rate of all agricultural machinery, the average operation efficiency of all subtasks, and the total operation cost of all subtasks under each scheme are multiplied by the corresponding evaluation index weights, and then the three results are added to obtain the comprehensive score of the scheme. The comprehensive scores of all schemes are summarized in the scheduling scheme comprehensive score table, which contains information such as scheme ID and comprehensive score.

[0192] Preferably, step S6 comprises the following steps:

[0193] Step S61: importing the resource scheduling plan into the agricultural machinery operation management system, generating agricultural machinery operation instructions, and obtaining agricultural machinery operation instructions; using the Internet of Things technology to monitor the agricultural machinery status and obtain the agricultural machinery status data stream;

[0194] Step S62: tracking the operation progress according to the agricultural machinery status data stream and the agricultural machinery operation instructions to obtain an operation progress report;

[0195] Step S63: performing abnormal event detection on the agricultural machinery status data stream to obtain abnormal event alarm data;

[0196] Step S64: acquiring real-time meteorological data; determining the dynamic scheduling triggering condition according to the abnormal event alarm data and the real-time meteorological data, and obtaining a dynamic scheduling triggering signal;

[0197] Step S65: Generate a dynamic scheduling solution using a multi-objective optimization model according to the dynamic scheduling trigger signal to obtain a dynamic scheduling solution;

[0198] Step S66: updating the agricultural machinery operation instruction according to the dynamic scheduling plan, and recording the operation data to obtain the operation record data; generating the agricultural machinery operation execution report according to the operation record data to obtain the agricultural machinery operation execution report.

[0199] In an embodiment of the present invention, the optimal resource scheduling scheme determined in step S5 is imported into the agricultural machinery operation management system. According to the allocation of each subtask in the scheme and the operation requirements, an agricultural machinery operation instruction containing information such as agricultural machinery ID, subtask ID, operation path, operation parameters, etc. is generated and sent to the corresponding agricultural machinery terminal through a wireless network. Using sensors deployed on agricultural machinery, such as GPS positioning sensors, oil volume sensors, engine speed sensors, etc., the operation status data of agricultural machinery, such as position information, fuel consumption information, operation speed, etc., are collected in real time. The collected data is uploaded to the agricultural machinery operation management system through a wireless network to form an agricultural machinery status data stream. According to the location information and operation time information in the agricultural machinery status data stream, combined with the preset operation path and operation time in the agricultural machinery operation instruction, the operation progress of each subtask is tracked in real time, such as the completed operation area, the remaining operation area, the estimated completion time, etc. The operation progress information of all subtasks is summarized to generate an operation progress report, and is visually displayed in the agricultural machinery operation management system. Real-time analysis of the agricultural machinery status data stream is performed. For example, an early warning threshold is set. When the oil volume of the agricultural machinery is lower than the early warning threshold, an insufficient oil volume alarm is triggered; when the operating speed of the agricultural machinery is lower than the preset normal range, an abnormal operating speed alarm is triggered. All detected abnormal event information is recorded in the abnormal event alarm data, and relevant personnel are notified in time for processing. Real-time meteorological data, such as rainfall, wind speed, temperature, etc., are obtained by calling the meteorological data interface. According to the preset dynamic scheduling trigger conditions, for example, when the rainfall exceeds the preset threshold, or the wind speed exceeds the safe operating range of the agricultural machinery, it is determined whether dynamic scheduling is required. If the dynamic scheduling trigger conditions are met, a dynamic scheduling trigger signal is generated. When the dynamic scheduling trigger signal is received, the existing resource scheduling scheme is adjusted according to the abnormal event alarm data and the real-time meteorological data. For example, if a certain agricultural machinery fails, the agricultural machinery is removed from the scheme, and the subtasks that the agricultural machinery is responsible for are reallocated to other available agricultural machinery. The adjusted scheme is recalculated using the multi-objective optimization model constructed in step S4 to obtain a new dynamic scheduling scheme. According to the dynamic scheduling scheme generated in step S65, the agricultural machinery operation instructions are updated and sent to the corresponding agricultural machinery terminal. At the same time, the actual operation data of the agricultural machinery, such as operation trajectory, operation time, fuel consumption, operation quality, etc., are recorded to form operation record data. According to the operation record data, an agricultural machinery operation execution report containing information such as operation completion status, operation efficiency, operation cost, abnormal event record, etc. is generated to provide data support for subsequent operation analysis and management.

[0200] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0201] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A multi-objective task optimization scheduling method based on agricultural machinery operation efficiency, characterized in that: The following steps are involved: Step S1: measuring the operating efficiency parameters of each agricultural machine under different operating conditions to obtain an agricultural machine operating efficiency data set; The agricultural machinery capacity model is built based on the agricultural machinery operation efficiency data set to obtain the agricultural machinery capacity matrix; Step S2: using remote sensing technology to partition the farmland and obtain a farmland partition map; decomposing the farmland operation tasks according to the farmland partition map, and generating subtask sets to obtain subtask sets; Step S3: generating collaborative edges according to the agricultural machinery capability matrix to obtain a collaborative edge set; performing collaborative edge cost weighting processing on the collaborative edge set according to the agricultural machinery capability matrix to obtain a weighted collaborative edge set; generating a collaborative relationship graph according to the weighted collaborative edge set to obtain a collaborative relationship graph; Step S4: construct a multi-objective optimization model according to the collaboration relationship diagram to obtain a multi-objective optimization model; perform Pareto solution processing according to the multi-objective optimization model to obtain a resource scheduling solution set; Step S5: using the simulation software and the resource scheduling solution set to run a simulation experiment and obtain simulation experiment result data; calculating the resource scheduling weight according to the simulation experiment result data, and generating a resource scheduling solution to obtain a resource scheduling solution; Step S6: Import the resource scheduling plan into the agricultural machinery operation management system, dynamically process the scheduling plan, and obtain a dynamic scheduling trigger signal; perform dynamic scheduling execution using a multi-objective optimization model based on the dynamic scheduling trigger signal to obtain an agricultural machinery operation execution report.

2. The multi-objective task optimization scheduling method based on agricultural machinery operation efficiency according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: collecting basic information of each agricultural machine to obtain an agricultural machine basic information table; measuring agricultural machine operation efficiency parameters under different operation conditions according to the agricultural machine basic information table to obtain an agricultural machine operation efficiency data set; Step S12: performing fuel consumption characteristic analysis based on the agricultural machinery operation efficiency data set and the agricultural machinery basic information table to obtain agricultural machinery fuel consumption characteristic data; Step S13: constructing an agricultural machinery fuel consumption model according to the agricultural machinery fuel consumption characteristic data to obtain an agricultural machinery fuel consumption model; Step S14: forecasting maintenance demand based on the agricultural machinery basic information table, the agricultural machinery operation efficiency data set, and the agricultural machinery fuel consumption model to obtain agricultural machinery maintenance demand planning data; Step S15: calibrating the agricultural machinery cooperation relationship parameters for each agricultural machinery to obtain an agricultural machinery cooperation relationship parameter table; Step S16: Generate an agricultural machinery capability matrix based on the agricultural machinery basic information table, the agricultural machinery operation efficiency data set, the agricultural machinery fuel consumption model, the agricultural machinery maintenance demand plan data, and the agricultural machinery cooperation relationship parameter table to obtain the agricultural machinery capability matrix.

3. The multi-objective task optimization scheduling method based on agricultural machinery operation efficiency according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: using remote sensing technology to collect farmland information and obtain a farmland geographic information database; Step S22: Acquire agricultural activity demand data; identify the operation type according to the agricultural activity demand data to obtain an operation type list; Step S23: partitioning the farmland space according to the farmland geographic information database and the operation type list to obtain a farmland partition map; Step S24: Decomposing farmland operation tasks according to the farmland zoning map and the farmland geographic information database to obtain a subtask spatial distribution map; Step S25: assigning subtask attributes according to the subtask spatial distribution diagram to obtain a subtask attribute table; generating a subtask set according to the subtask attribute table to obtain a subtask set.

4. The multi-objective task optimization scheduling method based on agricultural machinery operation efficiency according to claim 3 is characterized in that: Step S24 includes the following steps: Step S241: Estimating the workload according to the farmland zoning map to obtain a workload estimation table; Step S242: determining the operation time window for the farmland zoning map to obtain an operation time window table; Step S243: performing obstacle identification on the farmland partition map to obtain an obstacle information layer; Step S244: performing a traversability analysis based on the obstacle information layer and the farmland geographic information database to obtain a traversability analysis diagram; Step S245: Perform subtask spatial clustering analysis based on the farmland zoning map, the workload estimation table, the operation time window table, and the accessibility analysis diagram to obtain a subtask spatial distribution map.

5. The multi-objective task optimization scheduling method based on agricultural machinery operation efficiency according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: creating agricultural machinery nodes according to the agricultural machinery capability matrix to obtain an agricultural machinery node set; creating subtask nodes according to the subtask information in the subtask set to obtain a subtask node set; Step S32: performing agricultural machinery-subtask matching according to the subtask node set and the agricultural machinery node set to obtain an agricultural machinery-subtask matching table; Step S33: identifying the agricultural machinery-agricultural machinery cooperation relationship according to the agricultural machinery-subtask matching table and the agricultural machinery capability matrix, and obtaining the agricultural machinery-agricultural machinery cooperation relationship table; Step S34: generating collaborative edges according to the agricultural machinery-subtask matching table and the agricultural machinery-agricultural machinery collaborative relationship table to obtain a collaborative edge set; Step S35: performing collaborative edge cost weighting processing on the collaborative edge set according to the agricultural machinery-subtask matching table and the agricultural machinery-agricultural machinery collaborative relationship table to obtain a weighted collaborative edge set; Step S36: Generate a collaboration relationship graph based on the agricultural machinery node set, the subtask node set, and the weighted collaboration edge set to obtain a collaboration relationship graph.

6. The multi-objective task optimization scheduling method based on agricultural machinery operation efficiency according to claim 5 is characterized in that: Step S35 includes the following steps: Step S351: Calculate the time cost according to the agricultural machine-subtask matching table and the agricultural machine-agricultural machine cooperation relationship table to obtain a time cost matrix; Step S352: Calculate the fuel cost according to the agricultural machinery-subtask matching table and the agricultural machinery fuel consumption model in the agricultural machinery capability matrix to obtain a fuel cost matrix; Step S353: Perform weighted sum analysis on the time cost matrix and the fuel cost matrix to obtain a collaboration cost matrix; Step S354: assigning collaborative edge weights to the collaborative edge set according to the cost values ​​in the collaborative cost matrix to obtain a weighted collaborative edge set.

7. The multi-objective task optimization scheduling method based on agricultural machinery operation efficiency according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: define decision variables according to the collaboration diagram to obtain a decision variable set; construct an objective function according to the collaboration diagram to obtain an objective function expression; define constraint conditions according to the collaboration diagram to obtain a constraint condition set; Step S42: constructing a multi-objective optimization model according to the decision variable set, the objective function expression and the constraint condition set to obtain a multi-objective optimization model; Step S43: selecting a multi-objective evolutionary algorithm according to the multi-objective optimization model to obtain a selected multi-objective evolutionary algorithm; generating an initial population according to the selected multi-objective evolutionary algorithm and the multi-objective optimization model to obtain initial population data; Step S44: Calculate the fitness function according to the objective function expression to obtain a fitness value set; Step S45: using the selected multi-objective evolutionary algorithm to perform population evolution operations on the initial population data to obtain new population data; calculating the fitness value of the new population data to obtain the population fitness value; Step S46: Performing Pareto dominance relationship judgment on the population fitness values ​​according to the fitness value set to obtain a Pareto dominance relationship matrix; extracting resource scheduling solutions according to the Pareto dominance relationship matrix to obtain a resource scheduling solution set.

8. The multi-objective task optimization scheduling method based on agricultural machinery operation efficiency according to claim 1 is characterized in that: Step S5 includes the following steps: Step S51: constructing a farmland operation simulation environment using simulation software to obtain a virtual farmland operation environment; Step S52: importing the resource scheduling solution set into the virtual farmland operation environment, setting the agricultural machinery operation parameters, and obtaining the agricultural machinery operation simulation parameters; Step S53: performing simulation experiments according to the virtual farmland operation environment and the agricultural machinery operation simulation parameters to obtain simulation experiment result data; Step S54: Calculate resource scheduling weights based on simulation experiment result data to obtain a scheduling scheme comprehensive scoring table; Step S55: Generate a resource scheduling plan according to the scheduling plan comprehensive scoring table to obtain a resource scheduling plan.

9. The multi-objective task optimization scheduling method based on agricultural machinery operation efficiency according to claim 8 is characterized in that: Step S54 includes the following steps: Step S541: Calculate the utilization rate of each agricultural machine under each scheduling scheme according to the simulation experiment result data to obtain an agricultural machine utilization rate statistics table; Step S542: Calculate the operating efficiency of each subtask under each scheduling scheme according to the simulation experiment result data to obtain an operating efficiency statistics table; Step S543: Calculate the operation cost of each subtask under each scheduling scheme according to the simulation experiment result data to obtain an operation cost statistics table; Step S544: acquiring farmland operation demand data; determining the evaluation index weights according to the farmland operation demand data to obtain the evaluation index weight vector; Step S545: Calculate the comprehensive score of the scheduling scheme according to the agricultural machinery utilization rate statistics table, the operation efficiency statistics table, the operation cost statistics table and the evaluation index weight vector to obtain a scheduling scheme comprehensive score table.

10. The multi-objective task optimization scheduling method based on agricultural machinery operation efficiency according to claim 1 is characterized in that: Step S6 includes the following steps: Step S61: importing the resource scheduling plan into the agricultural machinery operation management system, generating agricultural machinery operation instructions, and obtaining agricultural machinery operation instructions; using the Internet of Things technology to monitor the agricultural machinery status and obtain the agricultural machinery status data stream; Step S62: tracking the operation progress according to the agricultural machinery status data stream and the agricultural machinery operation instructions to obtain an operation progress report; Step S63: performing abnormal event detection on the agricultural machinery status data stream to obtain abnormal event alarm data; Step S64: acquiring real-time meteorological data; determining the dynamic scheduling triggering condition according to the abnormal event alarm data and the real-time meteorological data, and obtaining a dynamic scheduling triggering signal; Step S65: Generate a dynamic scheduling solution using a multi-objective optimization model according to the dynamic scheduling trigger signal to obtain a dynamic scheduling solution; Step S66: updating the agricultural machinery operation instruction according to the dynamic scheduling plan, and recording the operation data to obtain the operation record data; generating the agricultural machinery operation execution report according to the operation record data to obtain the agricultural machinery operation execution report.

Images