Method for automatically distributing farming tasks based on multi-objective optimization model

Abstract

The invention relates to the technical field of big data analysis and processing, and is applied to the application of a multi-objective optimization model in the agricultural field, farmland management, crop planting and agricultural automation technologies. Factors such as farmland soil types, fertility conditions, moisture conditions, skills and experiences of farmers and the like are comprehensively considered through the multi-objective optimization model, the farmers and farming tasks are intelligently matched, and the accuracy of task allocation is improved. The process not only ensures that each task can be accurately allocated to the most appropriate executor, but also reduces resource waste caused by mismatching of the tasks and the executor ability, thereby improving the overall farming efficiency and effect. Secondly, according to the optimized farmland resource management scheme, optimal allocation of resources is realized by combining soil and environment information according to data such as the growth cycle, the water demand and the fertilizer demand of crops. Moreover, through automatic agricultural task allocation, the burden of the farmers on task allocation is remarkably relieved, the farmers can put more energy into actual agricultural operation and quality management, the risk of human decision errors is effectively reduced, and meanwhile, the overall efficiency and output quality of agricultural production are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis and processing, and is applied to the application of multi-objective optimization models in the agricultural field, farmland management, crop planting and agricultural automation technology. Background Art

[0002] Traditional farmland management relies on experienced farmers to perform agricultural activities such as irrigation, fertilization and sowing. However, this method is inefficient, wastes resources seriously, and may cause environmental pollution. With the development of the Internet of Things, big data and artificial intelligence technologies, the automated allocation of agricultural tasks has become possible.

[0003] A multi-objective optimization model is a mathematical model that aims to optimize multiple objective functions simultaneously. In the agricultural field, multi-objective optimization models can be used to maximize crop yields, minimize costs, and protect the environment at the same time. By using multi-objective optimization models, the automatic allocation of agricultural tasks can be achieved, thereby improving agricultural production efficiency, reducing costs, and optimizing resource allocation.

[0004] In order to realize the automatic allocation of agricultural tasks, it is necessary to collect and integrate farmer information, farmland information, crop growth cycle data and historical agricultural activity records. Farmer information includes data such as farmer's skill level, experience, and available time. Farmland information includes data such as soil type, fertility level, and moisture status. Crop growth cycle data includes growth period, water requirement, fertilizer requirement, etc. Historical agricultural activity records include the time, amount and method of irrigation, fertilization, sowing, etc.

[0005] Using multi-objective optimization models, we can automatically generate agricultural task allocation plans based on farmer skills, farmland conditions, and crop requirements. By automating agricultural equipment to perform predetermined agricultural tasks, we can improve agricultural production efficiency, reduce costs, and optimize resource allocation. At the same time, by real-time monitoring of farmland conditions and crop growth status, we can adjust the optimization model and task allocation strategy based on the implementation results, and further improve the automated allocation plan for agricultural tasks.

[0006] In short, with the development of the Internet of Things, big data and artificial intelligence technology, the application of multi-objective optimization models in the agricultural field has gradually attracted attention. By using multi-objective optimization models, we can realize the automated allocation of agricultural tasks, improve agricultural production efficiency, reduce costs, optimize resource allocation, and promote the sustainable development of agriculture. Summary of the invention

[0007] The core purpose of the present invention is to realize the automatic allocation of agricultural tasks through innovative technical solutions to improve the efficiency of agricultural production and the yield of crops. First, the present invention comprehensively considers factors such as farmland soil type, fertility status, water conditions, and farmers' skills and experience through a multi-objective optimization model, intelligently matches farmers with agricultural tasks, and improves the accuracy of task allocation. This process not only ensures that each task can be accurately allocated to the most suitable executor, but also reduces the waste of resources caused by the mismatch between tasks and executors' capabilities, thereby improving the overall agricultural efficiency and effect. Secondly, the optimized farmland resource management scheme of the present invention realizes the optimal allocation of resources based on data such as crop growth cycle, water demand, fertilizer demand, etc., combined with soil and environmental information. Furthermore, by automating the allocation of agricultural tasks, the present invention significantly reduces the burden of farmers on task allocation, enabling them to devote more energy to actual agricultural operations and quality management, effectively reducing the risk of human decision-making errors, and improving the overall efficiency and output quality of agricultural production.

[0008] The specific objectives can be broken down into the following aspects: (1) Improving the accuracy of agricultural task allocation: In traditional agricultural practice, the allocation of agricultural tasks often depends on the personal experience and subjective judgment of farmers, which may lead to a mismatch between tasks and farmers' skills and experience, resulting in waste of resources and inefficiency. The present invention uses a multi-objective optimization model to intelligently match farmers with agricultural tasks by deeply analyzing farmers' skill levels, historical agricultural execution data, and basic information such as soil type and fertility of farmland. This method not only significantly improves the accuracy of agricultural task allocation and ensures that each task can be allocated to the most suitable farmer, but also reduces excessive resource consumption and manpower waste caused by mismatched capabilities.

[0009] (2) Optimizing farmland resource management: Based on the growth cycle and specific needs of different crops (such as the length of the growing season, water and nutrient requirements), combined with data such as soil type, fertility and water conditions, the present invention can achieve more scientific and reasonable resource allocation. By carefully analyzing and systematically managing the farmland environment, the method of the present invention effectively improves the scientific nature and effectiveness of farmland management, thereby optimizing the overall growth environment of crops and providing a solid foundation for high and stable yields.

[0010] (3) Reducing farmers’ workload and labor intensity: The automated task allocation system relieves farmers of the tedious work of task allocation, greatly reducing their workload and physical burden. This improvement enables farmers to devote more energy to actual farming operations and crop quality management, while reducing the risks caused by human decision-making errors and improving the overall efficiency and quality of agricultural production.

[0011] (4) Improve crop yield and quality: The precision farming management scheme of the present invention ensures that crop growth conditions are optimized to the greatest extent, thereby promoting the healthy growth of crops and significantly improving crop yield and quality. Through the fine control of farming activities, the present invention helps to enhance the economic benefits of agriculture and improve the market competitiveness of agricultural products.

[0012] (5) Achieving sustainable development of agricultural production: In the face of the dual challenges of global environmental change and resource pressure, the scientific task allocation method of the present invention promotes the transformation of agricultural production methods to a more environmentally friendly and sustainable direction. By rationally utilizing agricultural resources and reducing the excessive use of fertilizers and pesticides, the present invention helps to reduce the negative impact on the environment and promote the green development of agricultural production.

[0013] (6) Improving the adaptability and flexibility of agricultural production: The present invention can quickly adjust the allocation of agricultural tasks based on feedback by real-time monitoring of farmland environment and crop growth conditions, combined with a flexible multi-objective optimization model. This dynamic adjustment mechanism enables agricultural production to better adapt to external environmental changes such as climate change and pest outbreaks, thereby improving the flexibility and resilience of agricultural production.

[0014] (7) Promote agricultural technological progress and knowledge popularization: The implementation of this invention not only promotes the development of agricultural information and intelligent technology, but also helps to spread and apply advanced agricultural knowledge and technology to a wider range of farmers through case demonstrations and technical training. The popularization of such knowledge and the dissemination of technology are conducive to improving the overall agricultural production level, accelerating the process of agricultural modernization, and laying the foundation for building an efficient and sustainable modern agricultural system.

[0015] Through the realization of the above-mentioned objectives, the present invention not only optimizes the allocation of agricultural resources and reduces the labor intensity of farmers by realizing the automatic allocation of agricultural tasks, but also promotes the improvement of crop yield and quality, thereby promoting the sustainable development and technological progress of agricultural production, enhancing the competitiveness of agricultural production, and promoting the harmonious development of agriculture and society.

[0016] The method for automatically allocating agricultural tasks based on a multi-objective optimization model includes the following steps: S1. Data collection and processing: Collecting farmers' skill text information: Collecting successful planting information through questionnaires or online inquiries, obtaining farmers' planting experience, professional skills text information and storing it in the database; Collecting historical farming data: Obtaining farmers' past farming activity records through database inquiries, and obtaining irrigation, fertilization, and sowing time information; Collecting basic farmland information: Obtaining soil type, fertility, and moisture status data through field visits and soil testing; Collecting crop growth cycle data: Obtaining crop growth requirements by observing and recording crop growth period, water demand, and fertilizer demand indicators; S2. Construction of multi-objective optimization model: S201. Determine the objective function According to the agricultural text information in the database, the information on maximizing crop yield and minimizing resource consumption is selected to construct the objective function of the multi-objective optimization problem; maximizing crop yield is measured by maximizing the economic yield of agricultural crops; minimizing resource consumption includes various resources required for agricultural production such as water, fertilizers, pesticides, and labor; S202, Quantitative Targets Maximizing crop yield uses historical data to select the highest possible yield under current farm conditions, and minimizing resource consumption adds up the expected consumption of each resource; S203. Determine constraints Set constraints for farm task allocation based on farmers’ skills, basic farmland information, and crop growth cycle data; S204. Weight Allocation In multi-objective optimization, different weights are set according to different objectives; S205. Constructing multi-objective optimization function Integrate multiple objectives into a single optimization function in a weighted manner; S206. The comprehensive objective function enters the data model, and the model algorithm is optimized using a genetic algorithm to obtain the optimal task allocation solution.

[0017] Compared with existing methods, the innovation of the method described in this paper lies in the use of a comprehensive multi-objective optimization model, which can more deeply consider a variety of planting environment factors including soil type, fertility status, water conditions and historical agricultural data. This all-round consideration effectively solves the problem of insufficient consideration of planting environment factors in the traditional decision-making process, and ensures that the allocation of agricultural tasks is more scientific and reasonable. Through this innovative method, not only the accuracy of decision-making is significantly improved, but also the quality of decision-making is improved, providing farmers with more accurate agricultural guidance, thereby fundamentally optimizing the agricultural production process, increasing crop yield and quality, and promoting the dual improvement of agricultural production efficiency and sustainability.

[0018] Technical points of the invention: 1. Construction and application of multi-objective optimization model One of the core points of this invention is to build a multi-objective optimization model, which comprehensively considers multiple objectives such as crop growth requirements, farmland resource limitations, environmental factors, etc., to achieve efficient and sustainable agricultural production. Technical points include: Determination of model input parameters: Collect and analyze crop growth data, soil characteristics, climate conditions, etc. to ensure the scientificity and accuracy of model input parameters. Setting of objective function: Take crop yield, resource utilization efficiency, environmental impact, etc. as objective functions to achieve multi-objective optimization of agricultural production. Establishment of constraint conditions: Consider the constraints of irrigation water sources, fertilizer supply, labor, etc. to ensure the feasibility of the optimization plan.

[0019] Model solving algorithm: Use advanced optimization algorithms (such as genetic algorithm, particle swarm optimization, etc.) to solve multi-objective optimization problems and obtain optimal or approximately optimal irrigation and fertilization plans.

[0020] 2. Development of Data Collection and Management Systems In order to support the effective operation of the multi-objective optimization model, the present invention has developed a data collection and management system. The technical points include: Data collection equipment: A sensor network deployed in farmland collects data such as soil moisture, temperature, and nutrient content in real time.

[0021] Data transmission and storage: Use wireless communication technology to transmit the collected data to the cloud server, and establish a database on the server to store and manage the data.

[0022] Data processing and analysis: Develop data processing software to pre-process, analyze and mine the collected data to provide input data for multi-objective optimization models.

[0023] Data visualization: Display farmland data and optimization results through the user interface to help farmers understand and make decisions.

[0024] 3. Design of automatic task allocation algorithm The present invention also relates to a set of automatic task allocation algorithms, which aims to realize the intelligent allocation of irrigation and fertilization tasks. The technical points include: Task Prioritization: Prioritize tasks based on crop growth stage, soil conditions, and resource availability.

[0025] Automated execution plan: Generates automated execution plans for irrigation and fertilization based on optimization models and task priorities, reducing human intervention.

[0026] Dynamic adjustment mechanism: Real-time monitoring of farmland status and task execution, and dynamic adjustment of task allocation as needed to cope with unforeseen environmental changes.

[0027] User interaction design: Provide farmers with an interactive platform that allows them to view task allocation and make manual adjustments when necessary.

[0028] Beneficial effects of the invention: The present invention brings the following beneficial effects through the implementation of the above technology: Improve agricultural production efficiency and crop yields: Maximize crop growth potential and increase agricultural output through scientific and reasonable irrigation and fertilization plans.

[0029] Reduce farmers’ workload and labor intensity: Automated task allocation and execution reduce farmers’ labor input on farmland and improve production convenience.

[0030] Optimize farmland resource allocation and management: Multi-objective optimization models help to rationally arrange irrigation and fertilization, improve resource utilization efficiency, and reduce waste.

[0031] Improve the accuracy of crop growth cycle monitoring and adjustment: Real-time data collection and analysis provide farmers with accurate crop growth information, helping to adjust management measures in a timely manner.

[0032] Environmentally friendly: Through precise irrigation and fertilization, the impact on the environment is reduced, which is in line with the development trend of green agriculture.

[0033] Increase farmers' income: Increasing crop yields and reducing production costs indirectly increase farmers' income levels.

[0034] Promoting agricultural modernization: The implementation of this invention will help promote the transformation of agricultural production methods towards modernization and intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a schematic diagram of the comparison before and after the allocation of agricultural tasks; Figure 2 It is a balance diagram between farmer workload and crop growth requirements; Figure 3 Automatically assign flow charts to farming tasks; Figure 4 Optimize the flow chart for the genetic algorithm. DETAILED DESCRIPTION

[0036] In order to make the above-mentioned objects, features and advantages of the present invention more understandable, the present invention is further described in detail below through detailed technical schemes and specific embodiments.

[0037] The system includes the following key steps: data collection, construction of a multi-objective optimization model, automated allocation of agricultural tasks, and real-time monitoring and feedback adjustment. The system automatically allocates the most suitable agricultural tasks based on the farmer's skill data and historical agricultural data, reducing resource waste and inefficiency caused by skill mismatch. Through the multi-objective optimization model, the system can intelligently allocate resources such as irrigation and fertilization, optimize resource consumption, and improve agricultural production efficiency. The system automatically allocates agricultural tasks, reducing the workload and labor intensity of farmers, allowing farmers to focus more on actual agricultural operations. The system provides real-time monitoring functions, and farmers can keep abreast of farmland conditions and crop growth through the user interface, and adjust task allocation based on feedback to optimize agricultural management.

[0038] The method for automatically allocating agricultural tasks based on a multi-objective optimization model is characterized by comprising the following steps: S1. Data collection and processing: Collecting farmers' skill text information: Collecting successful planting information through questionnaires or online inquiries, obtaining farmers' planting experience, professional skills text information and storing it in the database; Collecting historical farming data: Obtaining farmers' past farming activity records through database inquiries, and obtaining irrigation, fertilization, and sowing time information; Collecting basic farmland information: Obtaining soil type, fertility, and moisture status data through field visits and soil testing; Collecting crop growth cycle data: Obtaining crop growth requirements by observing and recording crop growth period, water demand, and fertilizer demand indicators; S2. Construction of multi-objective optimization model: S201. Determine the objective function According to the agricultural text information in the database, the information on maximizing crop yield and minimizing resource consumption is selected to construct the objective function of the multi-objective optimization problem; maximizing crop yield is measured by maximizing the economic yield of agricultural crops; minimizing resource consumption includes various resources required for agricultural production such as water, fertilizers, pesticides, and labor; S202, Quantitative Targets Maximizing crop yield uses historical data to select the highest possible yield under current farm conditions, and minimizing resource consumption adds up the expected consumption of each resource; S203. Determine constraints Set constraints for farm task allocation based on farmers’ skills, basic farmland information, and crop growth cycle data; S204. Weight Allocation In multi-objective optimization, different weights are set according to different objectives; S205. Constructing multi-objective optimization function Combine multiple objectives into a single optimization function in a weighted manner: F(x)=w1·xf1(x)+w2·f2(x) Among them, F(x) is the comprehensive objective function, f1(x) and f2(x) represent the objective functions of maximizing output and minimizing resource consumption, respectively, and w1 and w2 are the corresponding weights; S206. The comprehensive objective function enters the data model, and the model algorithm is optimized using a genetic algorithm to obtain the optimal task allocation solution.

[0039] The specific steps of genetic algorithm optimization in S206 are as follows: (1) Initialize the population: Randomly generate a certain number of comprehensive objective functions to form a population. Each comprehensive objective function represents a solution and consists of a certain number of field information. (2) Evaluation: The fitness of each comprehensive objective function in the population is evaluated, and the comprehensive objective function with high fitness is selected for reproduction; (3) Selection: Parent individuals are selected through roulette wheel and tournament selection operations based on the fitness of the comprehensive objective function; (4) Crossover: The genes of parent individuals are combined through the crossover operation to generate new offspring individuals; (5) Mutation: Perform mutation operations on newly generated offspring individuals, that is, randomly change the constraints to introduce new genetic diversity; Replacement: Replace some comprehensive objective functions in the population with the newly generated offspring individuals, thereby updating the population; (6) Termination condition: Repeat the above steps until the termination condition is met, the maximum number of reproduction iterations is reached, a sufficiently good solution is found, or the fitness no longer improves. The general form of the fitness function is: Among them, F(x) is the fitness, is the value of the ith objective function, n is the number of objective functions; The selection probability is calculated using the formula of the roulette selection algorithm: Where P(i) is the probability of selecting the i-th individual, f(i) is the fitness of the i-th individual, and N is the total number of individuals in the population.

[0040] The following are specific examples that illustrate how weights are applied in practice and influence the decision-making process.

[0041] Case background: Assume that a farm needs to grow wheat and corn with limited water resources. The goal is to maximize the total yield of the two crops while ensuring efficient use of water resources.

[0042] Weight setting: In this case, we set: w1=0.4: represents the weight of maximizing output. Because the market price is stable, farms are more concerned with maintaining a higher production level.

[0043] w2=0.6: represents the weight for minimizing water consumption, reflecting the scarcity and high cost of water resources.

[0044] Objective function definition: f1(x): Yield maximization function, defined as the sum of the expected yields of the two crops.

[0045] f2(x): Water consumption minimization function, defined as the sum of the water required to grow the two crops.

[0046] Optimization process: Optimization is performed through genetic algorithms, and each solution corresponds to a different crop planting ratio and water resource allocation plan.

[0047] Plan A: When w1=0.4 and w2=0.6, the optimization results show that the planting ratio of wheat and corn is 3:2, and water resources are mainly allocated to wheat because wheat has a higher water utilization efficiency.

[0048] Solution B: In contrast, if we adjust the weights to w1=0.6 and w2=0.4, the optimization result may show a planting ratio of 1:1. At this time, the water resource distribution is more balanced, but due to excessive use of water resources, it may lead to resource shortages.

[0049] Effect verification: Through field testing, after implementation, Option A showed that water usage was reduced by 25% compared to traditional methods, while total production only decreased by 5%. This shows that by properly adjusting the weights, more efficient resource utilization and acceptable production levels can be achieved when resources are limited.

[0050] Through this case, we can see that the allocation of weights w1 and w2 directly affects the decision-making of agricultural tasks and the optimal allocation of resources.

[0051] Genetic Algorithm (GA) is a search heuristic algorithm that simulates biological evolution in nature. It was proposed by Holland in 1965. It solves problems by simulating natural selection and gene manipulation in genetics, and is particularly suitable for complex optimization problems. The specific formulas of crossover and mutation operations depend on the crossover and mutation strategies adopted.

[0052] To fully evaluate the effectiveness of the genetic algorithm, a detailed experiment was designed to compare the performance of the genetic algorithm with the traditional method in the allocation of agricultural tasks. Different environmental conditions, crop types, and resource constraints were considered in the experiment.

[0053] Experimental group: Use genetic algorithm for task allocation.

[0054] Control group: Use the traditional empirical allocation method.

[0055] Experimental parameters: Result analysis: Resource consumption - Genetic algorithm group: average water consumption is 4,500 cubic meters per hectare. Traditional method group: average water consumption is 5,600 cubic meters per hectare. Percentage of savings: The genetic algorithm saves about 19.6% of water resources.

[0056] Yield - Genetic algorithm group: Average yield was 5.8 tons per hectare. Traditional method group: Average yield was 5.1 tons per hectare.

[0057] Percentage increase in yield: The genetic algorithm increased the yield by about 13.7%.

[0058] Cost-Benefit Analysis: Cost - Genetic Algorithm Group: Due to the optimization of resource allocation, the use of water resources and labor is reduced, and the cost is relatively low. Traditional Method Group: Uneven use of resources leads to high costs. Benefits - Genetic Algorithm Group: Due to the increase in output, the benefits are relatively increased. Traditional Method Group: The output is low and the benefits are limited. Environmental Impact - Genetic Algorithm Group: Due to the reduction in the use of water resources, the negative impact on the environment is small. Traditional Method Group: Excessive use of water resources may lead to a drop in groundwater levels and damage to the ecological environment.

[0059] in conclusion: Genetic algorithms perform well in the automatic allocation of agricultural tasks, not only significantly increasing crop yields, but also effectively reducing resource consumption and the negative impact of agricultural production on the environment. Through this optimization, agricultural production can be more sustainable while improving economic benefits. Through this detailed effect comparison analysis, we can see the obvious advantages of genetic algorithms in practical applications. This method not only provides an efficient resource management tool for agricultural production, but also promotes the development and innovation of agricultural technology.

[0060] The following is part of the algorithm code: import numpy as np # Assume we have two objective functions f1(x) = x^2 and f2(x) = (x-2)^2 def evaluate(individual): x = individual[0] f1 = x**2 f2 = (x - 2) ** 2 return f1, f2 # Genetic Algorithm Class class GeneticAlgorithm: def __init__(self, pop_size=100, num_generations=100, crossover_rate=0.8, mutation_rate=0.01): self.pop_size = pop_size self.num_generations = num_generations self.crossover_rate = crossover_rate self.mutation_rate = mutation_rate self.population = np.random.rand(pop_size, 1) # Initialize the population, which only contains one gene def evolve(self): for generation in range(self.num_generations): # Evaluate fitnesses = [evaluate(ind) for ind in self.population] # Sort or select multiple targets based on fitnesses # Selection, crossover, mutation (only pseudo code logic is shown here) new_population = [] for _ in range(self.pop_size / / 2): # Select two individuals parent1, parent2 = select_parents(self.population, fitnesses) # Cross if np.random.rand() <self.crossover_rate: child1, child2 = crossover(parent1, parent2) new_population.extend([child1, child2]) else: new_population.extend([parent1, parent2]) # Mutation (for newly generated population) for individual in new_population: if np.random.rand() <self.mutation_rate: individual = mutate(individual) # Update population self.population = new_population # Output the optimal solution (here it is assumed that two objective functions are minimized) best_index = np.argmin([sum(fitness) for fitness in fitnesses]) best_solution = self.population[best_index] return best_solution Automated distribution of agricultural tasks Based on the results of the multi-objective optimization model, combined with farmers' skills and historical farming data, farming tasks are automatically generated and assigned. Task assignment rules are established, such as irrigation time, fertilizer application amount, etc., based on crop growth requirements and basic farmland information.

[0061] Irrigation time rules: Monitor soil moisture: Use soil moisture sensors or remote sensing technology to monitor soil moisture, and trigger irrigation when soil moisture drops to a certain threshold. Consider crop water requirements: Arrange irrigation time according to the water requirements of different crops at different growth stages. For example, rice requires more water than wheat. Climate factors: Consider local climate conditions such as temperature, humidity, wind speed, etc. to determine the best time for irrigation to avoid evaporation losses caused by irrigation during high temperature and drought periods. Efficient use of water resources: Give priority to water-saving irrigation technologies such as drip irrigation and sprinkler irrigation, and arrange the irrigation sequence and time according to the specific conditions of the farmland to reduce water waste. Preventive irrigation: Predict drought conditions based on weather forecasts and climate data, and irrigate in advance to avoid crop damage due to lack of water.

[0062] Fertilizer dosage rules: Soil testing: Analyze the soil composition regularly to understand the soil fertility, especially the content of major nutrients such as nitrogen, phosphorus, and potassium. Crop needs: Determine the amount of fertilizer based on the specific needs of the crop. Different crops have different nutrient requirements, and the same crop has different requirements at different growth stages. Combination of organic and inorganic fertilizers: Use organic fertilizers and chemical fertilizers in combination to achieve long-term stability of soil fertility and healthy growth of crops. Environmental considerations: Avoid excessive fertilization that leads to nutrient loss and water pollution, while reducing negative impacts on the environment. Timing of fertilization: Reasonably arrange the timing of fertilization based on the crop growth cycle and soil nutrient conditions, and usually increase the amount of fertilizer during the critical period of crop growth.

[0063] Specific implementation: Data collection - Soil moisture data: collected every 30 minutes, using soil moisture sensors. Meteorological data: including temperature, precipitation, wind speed, etc., updated daily. Crop growth data: record crop types, growth stages and biomass. Water resource data: monitor available water and consumption. Farmer skill data: record farmers' professional skills and experience. Historical farming data: including past irrigation and fertilization records, etc.

[0064] Data preprocessing - Cleaning: Remove outliers and erroneous data. Standardization: Convert all data into a unified format and measurement unit. Integration: Integrate data from different sources into one database.

[0065] Construct a multi-objective optimization model - Objective function: maximize yield and minimize resource consumption. Constraints: consider soil moisture, meteorological conditions, crop requirements and farmer skills. Optimization algorithm: use genetic algorithm for optimization.

[0066] Farming task generation and allocation - Task generation: Generate specific irrigation and fertilization plans based on the results of the optimization model.

[0067] Task allocation: Irrigation tasks: Based on soil moisture and meteorological data, they are allocated to farmers with irrigation experience. Fertilization tasks: Based on soil fertility and crop requirements, they are allocated to farmers with fertilization experience.

[0068] Farming task execution - Execution monitoring: Real-time monitoring of task execution to ensure that it is carried out as planned. Execution feedback: Collect data during the execution process, such as actual water usage and fertilizer application.

[0069] Feedback adjustment - data analysis: Analyze the difference between the execution data and the expected results. Model adjustment: Adjust the parameters of the optimization model according to the feedback. Task adjustment: Adjust the subsequent task allocation according to the new model results.

[0070] Embodiment 1: Suppose there is a 10-hectare wheat field managed by 5 farmers. Through a questionnaire survey, we collect information about the farmers' planting experience, professional skills, etc. Each farmer has different skills and experience, as shown below: Farmer A: has 10 years of irrigation experience and is good at using water-saving technology. Farmer B: has 8 years of fertilization experience and has a deep understanding of soil fertility adjustment. Farmer C: newcomer, 1 year of experience, currently receiving training. Farmer D: has 15 years of agricultural experience and is good at crop pest management. Farmer E: has 5 years of experience and is good at data recording and analysis.

[0071] According to the results of the multi-objective optimization model, we get the following farming task plan: Irrigation plan: In the next 7 days, irrigation will be carried out in the early morning every day, and the irrigation volume is 500 cubic meters per hectare each time. Fertilization plan: Fertilization will be carried out on the 3rd and 6th days, with 20 kg / hectare of nitrogen fertilizer and 15 kg / hectare of phosphorus fertilizer applied each time.

[0072] Farming task allocation - irrigation task: Farmer A: responsible for the irrigation task for the first 4 days, using his experience in water-saving technology. Farmer C: cooperate with farmer A to learn irrigation technology. Fertilization task: Farmer B: responsible for the fertilization task, using his rich experience in fertilization. Farmer E: assist farmer B and be responsible for recording fertilization data. Farming task execution: Farmers A and C perform irrigation according to the plan, monitor soil moisture every day to ensure that the irrigation amount is appropriate. Farmers B and E fertilize according to the plan, and E is responsible for recording data such as fertilization amount and fertilization time.

[0073] Feedback adjustment - Farmer E inputs the collected data into the system, including actual water usage, fertilizer application, and any problems encountered. The system analyzes the data and finds that the amount of fertilizer applied on day 3 was higher than expected, resulting in slight fertilizer loss. Based on this feedback, the fertilization plan is adjusted to reduce the amount of fertilizer applied on day 6 to avoid further loss.

[0074] Result analysis: Through this automated allocation and execution, the irrigation and fertilization tasks of the wheat field were precisely controlled. The cooperation between farmers A and C improved irrigation efficiency and reduced water waste. The cooperation between farmers B and E ensured the scientific nature of fertilization and accurate data recording. In the end, wheat yield increased by 15% compared to last year, and resource utilization efficiency was significantly improved.

[0075] This perfect implementation not only demonstrates the automated allocation of farming tasks, but also takes into account the skills and experience of farmers, ensuring efficient execution of tasks and rational use of resources.

[0076] Feedback adjustment - farmer feedback and task adjustment: Farmers can provide feedback on the completion of farming tasks and crop growth conditions through the user interface, and the system adjusts task allocation based on the feedback.

[0077] Embodiment 2: Taking wheat planting as an example, the automated allocation process of agricultural tasks is described in detail: Data collection and processing: Through questionnaire surveys, collect information such as farmers' planting experience and professional skills. For example, farmer Zhang has 5 years of wheat planting experience and is good at irrigation and fertilization. Query the historical database to obtain records of farmer Zhang's past wheat planting activities, such as 3 irrigations and 2 fertilizations, as well as the results of each activity. Conduct field surveys of farmland to obtain data such as soil type, fertility, and moisture conditions. For example, the soil is loam, with medium fertility and appropriate moisture. Observe and record the growth cycle of wheat to obtain indicators such as growth period, water requirement, and fertilizer requirement. For example, the wheat growth period is 100 days, the water requirement is 100 cubic meters per mu, and the fertilizer requirement is 50 kg of nitrogen fertilizer per mu.

[0078] Construction of multi-objective optimization model: Determine the objective function: maximize wheat yield and minimize resource consumption. Determine the constraints: farmer Zhang San’s skills, basic information of farmland, wheat growth cycle data, etc. Select the optimization algorithm: Use genetic algorithm to solve multi-objective optimization problems and obtain the optimal task allocation solution.

[0079] Automatic allocation of agricultural tasks: Based on the results of the multi-objective optimization model, combined with farmer Zhang's skills and historical agricultural data, agricultural tasks are automatically generated and allocated. Establish task allocation rules: According to the water demand of wheat and the moisture status of the soil, the irrigation time is determined; according to the fertilizer demand of wheat and the fertility status of the farmland, the fertilizer amount is determined.

[0080] Adjusting task allocation: According to farmer Zhang’s workload and wheat growth requirements, the algorithm adjusts task allocation to balance workload and wheat growth requirements. Farmers perform agricultural tasks - According to the assigned agricultural tasks, farmers perform agricultural tasks and provide feedback on the execution of agricultural tasks. Feedback adjustment - farmer Zhang’s feedback and task adjustment: Zhang provides feedback on the completion of irrigation and fertilization, as well as the growth status of wheat, through the user interface, and the system adjusts task allocation based on the feedback.

[0081] Data analysis - In order to show the changes in resource consumption and crop yields before and after the optimization of agricultural task allocation, to show the changes in farmer workload and crop growth requirements over time, and to show the balance between them. The following is a graphical analysis of the results of the embodiment. Figure 1 This is a schematic diagram of the comparison before and after the allocation of agricultural tasks; Figure 2 This is a balance diagram between farmers' workload and crop growth requirements. Figure 2 Note: Growth requirements are based on key factors that affect crop growth requirements, such as water, nutrients, light, etc., and refer to other materials [1] Determine the sowing period as 5, with medium growth requirements, the growing period as 8, with high growth requirements, and the harvesting period as 2, with low growth requirements.

[0082] Technical features of the invention: Construction and application of multi-objective optimization model - integrated multi-source data input: This invention ensures the comprehensiveness and accuracy of model input by integrating multi-source information such as soil moisture, meteorological data, and crop growth data. For example, soil moisture sensors and satellite meteorological data are used to improve data accuracy. There are also data collection methods such as questionnaires and historical data retrieval.

[0083] Dynamic objective function setting: The model can dynamically adjust the objective function according to different crops and environmental conditions, such as paying more attention to water conservation in arid areas. This dynamic adjustment has been verified by experimental data, showing that it improves resource utilization efficiency by 20% compared with the traditional fixed weight model.

[0084] Development of data collection and management system - efficient data transmission mechanism: A set of IoT device docking APIs based on LoRaWAN has been developed, which can extract device data regularly for various types of agricultural IoT devices from different manufacturers and with different protocols using scheduled tasks, improving stability by 40%.

[0085] Intelligent data processing: Introducing machine learning technology (the company's self-developed Shennong No. 1 large model) for data analysis, which can predict soil trends, etc. In actual applications, the machine learning model accurately predicted 90% of soil moisture trends.

[0086] Interactive Data Visualization: We designed a web-based interactive data visualization system that allows farmers to view real-time data and historical trends directly on their mobile phones or computers. User satisfaction surveys show that ease of use has increased by 50%.

[0087] Design of automatic task allocation algorithm Intelligent adjustment of priorities: The algorithm can automatically adjust task priorities based on real-time data and weather forecasts, such as delaying irrigation if rain is forecast. During testing, this intelligent adjustment reduced unnecessary irrigation by 10%.

[0088] User interaction design: Provides a role-based access control that allows farmers to manually adjust tasks according to their needs, enhancing the flexibility of the system and user control. User feedback shows that this design improves their trust and acceptance of technology.

[0089] This patented solution can achieve: 1. Multi-objective optimization of agricultural production: By constructing a multi-objective optimization model, this patented solution can formulate scientific and reasonable irrigation and fertilization plans based on crop growth requirements, farmland resource limitations and environmental factors, so as to achieve efficient and sustainable agricultural production.

[0090] 2. Farmland data collection: This patented solution covers the entire farmland area by deploying IoT sensors. The gateway device transmits the collected data to the cloud server and cloud database via the Internet, realizing real-time collection, transmission, storage and management of farmland data, and providing data support for the multi-objective optimization model.

[0091] 3. Data visualization: Design a web-based interactive data visualization system so that farmers can view real-time data and historical trends through their mobile phones or computers. Provide a variety of visualization forms such as charts and maps to help farmers intuitively understand farmland conditions and optimization results.

[0092] 4. Automatic agricultural task allocation algorithm: The automatic task allocation algorithm designed by this patent solution can intelligently allocate irrigation and fertilization tasks according to crop growth stage, soil conditions and resource availability, thereby improving agricultural production efficiency.

[0093] Possible application scenarios in the future: 1. Intelligent farm management: In large farms or agricultural parks, the application of the system of the present invention can realize intelligent management of agricultural tasks, reduce human intervention, and improve agricultural production efficiency and crop yields.

[0094] 2. Precision agriculture practice: Combined with GPS positioning and remote sensing technology, the system of the present invention can be applied to the field of precision agriculture to achieve refined management of farmland and accurate allocation of resources.

[0095] 3. Agricultural cooperatives and family farms: For agricultural cooperatives and family farms, the system of the present invention can help them optimize resource allocation, improve production efficiency and reduce production costs.

[0096] 4. Agricultural research and teaching: In agricultural research institutions and colleges and universities, the system of the present invention can be used as a research tool and teaching case to explore and verify new agricultural technologies and management methods.

[0097] 5. Environmental monitoring and protection: By real-time monitoring of farmland environment and crop growth conditions, the system of the present invention helps to timely discover and solve environmental problems and promote sustainable agricultural development.

[0098] Glossary Multi-objective optimization model: A mathematical model that aims to optimize multiple objective functions simultaneously, in this scenario, the objectives include maximizing crop yields, minimizing costs, and protecting the environment.

[0099] Genetic Algorithm (GA): A search heuristic algorithm that simulates biological evolution in nature, proposed by Holland in 1965.

[0100] Agricultural tasks: refers to various activities in farmland management, such as irrigation, fertilization, sowing, etc.

[0101] Farmers: refers to the owners or managers of farmland, who are responsible for carrying out agricultural tasks.

Claims

1. A method for automatically allocating agricultural tasks based on a multi-objective optimization model, characterized in that The steps include: S1. Data collection and processing: Collecting farmers' skill text information: Collecting successful planting information through questionnaires or online inquiries, obtaining farmers' planting experience, professional skills text information and storing it in the database; Collecting historical farming data: Obtaining farmers' past farming activity records through database inquiries, and obtaining irrigation, fertilization, and sowing time information; Collect basic farmland information: obtain soil type, fertility, and moisture status data through field visits and soil testing; collect crop growth cycle data: obtain crop growth requirements by observing and recording crop growth period, water demand, and fertilizer demand indicators; S2. Construction of multi-objective optimization model: S201. Determine the objective function According to the agricultural text information in the database, the information on maximizing crop yield and minimizing resource consumption is selected to construct the objective function of the multi-objective optimization problem; maximizing crop yield is measured by maximizing the economic yield of agricultural crops; minimizing resource consumption includes various resources required for agricultural production such as water, fertilizers, pesticides, and labor; S202, Quantitative Targets Maximizing crop yield uses historical data to select the highest possible yield under current farm conditions, and minimizing resource consumption adds up the expected consumption of each resource; S203. Determine constraints Set constraints for farm task allocation based on farmers’ skills, basic farmland information, and crop growth cycle data; S204. Weight Allocation In multi-objective optimization, different weights are set according to different objectives; S205. Constructing multi-objective optimization function Combine multiple objectives into a single optimization function in a weighted manner: F(x)=w1·xf1(x)+w2·f2(x) Among them, F(x) is the comprehensive objective function, f1(x) and f2(x) represent the objective functions of maximizing output and minimizing resource consumption, respectively, and w1 and w2 are the corresponding weights; S206, the comprehensive objective function enters the data model, and the model algorithm is optimized using a genetic algorithm to obtain the optimal task allocation solution; The specific steps of genetic algorithm optimization in S206 are as follows: Initialize the population: randomly generate a certain number of comprehensive objective functions to form a population. Each comprehensive objective function represents a solution and consists of a certain number of field information. Evaluation: Evaluate the fitness of each comprehensive objective function in the population, and select the comprehensive objective function with high fitness for reproduction; Selection: Parent individuals are selected through roulette wheel and tournament selection operations according to the fitness of the comprehensive objective function; Crossover: Combine the genes of parent individuals through crossover operation to generate new offspring individuals; Mutation: Perform mutation operations on newly generated offspring individuals, that is, randomly change the constraints to introduce new genetic diversity; Replacement: Replace some comprehensive objective functions in the population with the newly generated offspring individuals to update the population; Termination condition: Repeat the above steps until the termination condition is met, the maximum number of reproduction iterations is reached, a sufficiently good solution is found, or the fitness no longer improves. The general form of the fitness function is: Among them, F(x) is the fitness, is the value of the ith objective function, n is the number of objective functions; The selection probability is calculated using the formula of the roulette selection algorithm: Where P(i) is the probability of selecting the i-th individual, f(i) is the fitness of the i-th individual, and N is the total number of individuals in the population.

Images