Intelligent agricultural planting management method and system based on AI analysis

Abstract

The invention relates to the technical field of intelligent agriculture, and discloses an intelligent agricultural planting management method and system based on AI analysis, and the method comprises the following steps: collecting environment data and market dynamics in real time, and carrying out data cleaning and data mapping output through edge nodes; uploading the processed data to a block chain network, generating a timestamp, and encrypting and storing the timestamp; and the interference income model calls block chain data to predict crop demands in the future 3-7 weeks and extra maintenance cost required by environment change, so as to generate an optimization scheme giving consideration to the maximum income and the minimum resource consumption. According to the method for predicting the crop demand and the maintenance cost required by the environmental change in the future 3-7 weeks through the interference income model, the response speed of agricultural management to sudden environmental change or market fluctuation is remarkably improved, the hysteresis of traditional manual decision making is avoided, and the precision and automation of dynamic resource allocation are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of smart agriculture, and particularly to an intelligent agricultural planting management method and system based on AI analysis. Background Art

[0002] In the current field of smart agriculture, traditional planting management methods mostly rely on manual experience and static data analysis, making it difficult to respond to sudden environmental changes or market fluctuations in real time, resulting in lagged decision-making and resource waste. In the prior art, the collection and processing of environmental data are often affected by sensor noise interference, reducing reliability, and the data storage lacks security and traceability, making it difficult to ensure information transparency and multi-party cooperation trust.

[0003] The prior patent discloses a smart agriculture production data decision management system based on AI technology (publication number CN115984028A), including an AI decision management platform. The AI decision management platform is communicatively connected to an agricultural supervision terminal. The AI decision management platform includes a server, and the server is communicatively connected to a data storage module, a regional environment monitoring module, a soil environment monitoring module, a multi-source data fusion evaluation module, and a decision management and warning module. The optimization model in the technology disclosed by this patent focuses on single-objective optimization and lacks a dynamic balance strategy that takes into account both benefits and resource consumption. Summary of the Invention

[0004] The present invention provides an intelligent agricultural planting management method and system based on AI analysis to solve the existing technical problems, and solves the problem of management lag caused by the inability to respond to sudden environmental changes or market fluctuations in real time.

[0005] To solve the above technical problems, according to one aspect of the present invention, more specifically, an intelligent agricultural planting management method based on AI analysis includes the following steps: S1. Real-time collect environmental data and market dynamics, and perform data cleaning and data mapping output through edge nodes; S2. Upload the processed data to the blockchain network, generate a timestamp and encrypt and store it; S3. The interference revenue model calls blockchain data to predict the crop demand and additional maintenance costs required for environmental changes in the next 3-7 weeks, and generates an optimization plan that takes into account the maximum benefit and the minimum resource consumption; S4. Cross-regionally collaborate with multiple platforms to analyze the resource redundancy of each farm, and match resource demanders and suppliers through a price priority strategy; S5. The execution terminal adjusts the parameters of agricultural planting equipment according to the optimization plan, and at the same time feeds back the operation results to the blockchain network to form a closed-loop optimization.

[0006] Furthermore, the intelligent agricultural planting management method based on AI analysis further includes: 1), Obtain images of the plant growth status based on drone inspection images; 2), Extract the characteristics of the plant growth status according to the image division; 3), Directly adjust the parameters of agricultural planting equipment according to the characteristics of the plant growth status.

[0007] Furthermore, the characteristics of the plant growth status include the leaf area index and pest and disease spots.

[0008] Furthermore, the environmental data is the soil temperature and humidity, and the soil nitrogen, phosphorus, and potassium content obtained based on environmental sensors; The market dynamics are the real-time prices and demand of crops obtained based on market data interfaces.

[0009] Furthermore, the data cleaning eliminates environmental sensor noise through Kalman filtering; The data mapping output calculates the additional maintenance costs required due to environmental changes in the next 3 - 7 weeks based on the soil temperature and humidity, and the soil nitrogen, phosphorus, and potassium content.

[0010] Furthermore, the interference benefit model calculates the cost-performance coefficient for obtaining the maximum benefit under the condition of minimum resource consumption through the actual benefit, theoretical benefit, and minimum resource consumption. Its calculation formula is: ; In the formula, represents the cost-performance coefficient for obtaining the maximum benefit under the condition of minimum resource consumption; represents the actual additional benefit that the crops in the next 3 - 7 weeks can obtain, in ten thousand yuan per hectare; represents the theoretical additional benefit that the crops in the next 3 - 7 weeks can obtain, in ten thousand yuan per hectare; and there is: ; In the formula, represents the minimum additional resource consumption required due to environmental changes in the next 3 - 7 weeks, in ten thousand yuan per hectare.

[0011] Furthermore, the interference benefit model calculates the cost-performance coefficient for the data according to the collected data, ranks the data according to the size of the cost-performance coefficient, and adjusts the parameters of agricultural planting equipment for the data corresponding to the maximum cost-performance coefficient.

[0012] Furthermore, the price priority strategy means ranking by the bids of resource demanders and preferentially delivering crops to the dealer with the highest bid.

[0013] An intelligent agricultural planting management system based on AI analysis includes: The data acquisition module monitors soil temperature and humidity, and light intensity based on the deployed environmental sensors; and obtains real-time prices and demand quantities of agricultural crops according to the market data interface. The UAV inspection unit obtains the plant growth state characteristics based on the UAV inspection images. The interference revenue model generates planting plans and resource allocation schemes based on environmental data and market dynamics. The execution control terminal is used to connect the irrigation system, intelligent greenhouse, and UAV equipment, and automatically adjusts the operation parameters according to the policy instructions.

[0014] An intelligent agricultural planting management method and system based on AI analysis provided by the present invention, compared with the prior art, the effects achieved by this method are as follows: 1. The present invention collects and processes environmental data and market dynamics in real time through edge nodes, combines the Kalman filtering technology to eliminate sensor noise, improves data quality, and predicts the crop demand and the maintenance cost required for environmental changes in the next 3 - 7 weeks through the interference revenue model, and generates an optimization scheme that takes into account the maximum revenue and the minimum resource consumption.

[0015] 2. The method of predicting the crop demand and the maintenance cost required for environmental changes in the next 3 - 7 weeks through the interference revenue model of the present invention significantly improves the response speed of agricultural management to sudden environmental changes or market fluctuations, avoids the lag of traditional manual decision-making, and realizes the precision and automation of dynamic resource allocation.

[0016] 3. The present invention encrypts and stores the cleaned data through blockchain technology and attaches time stamps to ensure the immutability, timeliness, and transparency of the data. The entire process from data acquisition to execution feedback is recorded through blockchain, providing credible data traceability support for farmers, dealers, and regulatory departments, enhancing the collaborative trust of the agricultural supply chain, and at the same time providing a reliable basis for dispute resolution. Description of the Drawings

[0017] Figure 1 is the flow chart of the intelligent agricultural planting management method in the present invention; Figure 2 is the theoretical additional revenue in the present invention and the minimum resource consumption relation diagram; Figure 3 is the relation diagram of the cost performance coefficient c, the actual additional revenue m, and the minimum resource consumption a in the present invention; Figure 4 is the predicted curve graph of the maintenance cost in the next 3 - 7 weeks in the present invention; Figure 5 is the cost performance coefficient c in the present invention varying with the minimum resource consumption Trend chart of changes; Figure 6 Schematic diagram for detecting leaf area index of different farmlands in the present invention; Figure 7 Schematic diagram for detecting disease density of different farmlands in the present invention. Detailed implementation manners

[0018] To make the technical solution of the present invention clearer, the following further describes the present invention in detail with reference to the accompanying drawings and specific embodiments.

[0019] Embodiment 1 As Figure 1 、 4 shown, according to one aspect of the present invention, an intelligent agricultural planting management method based on AI analysis is provided, which collects environmental data and market dynamics in real time, and outputs data cleaning and data mapping through edge nodes; the environmental data is the soil temperature and humidity, and the soil nitrogen, phosphorus and potassium content obtained based on environmental sensors; the market dynamics are the real-time prices and demand quantities of agricultural crops obtained based on market data interfaces.

[0020] The edge node refers to a computing device close to the data source (such as a farmland sensor gateway), which is used for local data processing to reduce the cloud transmission delay.

[0021] Data cleaning is to remove noise or outliers in sensor data through algorithms (such as Kalman filtering) to improve data quality. Kalman filtering is a recursive algorithm that eliminates random errors in time series data through prediction and correction.

[0022] Data mapping is the process of converting raw data (such as soil humidity) into business metrics (such as maintenance cost).

[0023] Data mapping output is to calculate the additional maintenance cost required due to environmental changes in the next 3-7 weeks based on the soil temperature and humidity, and the soil nitrogen, phosphorus and potassium content. Then, calculate the additional maintenance cost in the next 3-7 weeks according to the soil temperature and humidity, and the content of nitrogen, phosphorus and potassium (as Figure 4 shown).

[0024] 1), where the cost of irrigation is: ; In the formula, represents the cost of irrigation; represents the water requirement of the agricultural crop (unit: mm / day), which is positively correlated with temperature and negatively correlated with humidity; represents the farmland area (hectare); represents the number of days (3-7 weeks); represents the irrigation efficiency (0.9 for drip irrigation and 0.6 for flood irrigation); represents the water price (unit: yuan per cubic meter). Among them, the calculation formula for the water requirement of crops is: ; In the formula, represents the reference evaporation; represents the crop coefficient; represents the base water requirement of the crop; represents the soil humidity; represents the correction intensity of the water requirement when the water requirement of the crop deviates from the base water requirement of the crop; represents the inhibition effect intensity of soil humidity on the water requirement.

[0025] 2) Among them, the cost of fertilization is: ; In the formula, represents the cost of fertilization; represents the number of days (3 - 7 weeks); represents the daily average demand of the crop for nitrogen / phosphorus / potassium on the i-th day (kg / hectare / day); represents the available nitrogen / phosphorus / potassium content in the soil on the i-th day (kg / hectare); represents the fertilizer utilization rate on the i-th day (for example, the nitrogen utilization rate of urea is 30% - 50%); represents the unit price of the fertilizer on the i-th day (yuan / kg).

[0026] Example 2 As Figure 1 shown, the processed data is uploaded to the blockchain network to generate a timestamp and be encrypted and stored. After the data is cleaned and mapped by the edge node (such as converting environmental data into maintenance cost prediction), it is uploaded to the blockchain network through the blockchain node interface. When uploading, the blockchain network generates a unique timestamp for each piece of data to ensure the timeliness and immutability of the data.

[0027] Example 3 As Figures 1-3 shown in 5, the interference revenue model calls blockchain data (the interference revenue model is a model locally deployed by combining market and environmental data, used to optimize the balance between resource input and revenue) to predict the additional maintenance costs required for crop demand and environmental changes in the next 3 - 7 weeks, and to generate an optimization plan that takes into account the maximum revenue and the minimum resource consumption. Among them, the interference revenue model calculates the cost - performance coefficient for obtaining the maximum revenue under the condition of minimum resource consumption through the actual revenue, theoretical revenue, and minimum resource consumption, and its calculation formula is: ; In the formula, represents the cost - performance coefficient for obtaining the maximum revenue under the condition of minimum resource consumption (such asFigure 3 as shown in represents the actual additional income that can be obtained by crops in the next 3 - 7 weeks, in ten thousand yuan per hectare; represents the theoretical additional income that can be obtained by crops in the next 3 - 7 weeks, in ten thousand yuan per hectare; and there is: ; In the formula, represents the minimum additional resource consumption required due to environmental changes in the next 3 - 7 weeks, in ten thousand yuan per hectare.

[0028] Among them, the minimum additional resource consumption required due to environmental changes in the next 3 - 7 weeks represents the sum of fertilization cost and irrigation cost ( ). And when the minimum additional resource consumption required per hectare of farmland in the next 5 weeks takes (ten thousand yuan), then the theoretical additional income that can be obtained by crops in the next 5 weeks is: ; If the actual additional income that can be obtained by crops in the next 5 weeks is (ten thousand yuan), then the cost - performance ratio coefficient for obtaining the maximum income under the condition of minimum resource consumption is: ; Based on the above calculations, it can be known that when the minimum additional resource consumption in the next 5 weeks takes (ten thousand yuan), the cost - performance ratio coefficient for obtaining the maximum income is . The interference income model calculates the cost - performance ratio coefficient under this data based on the collected data, ranks them according to the size of this cost - performance ratio coefficient, and adjusts the parameters of agricultural planting equipment for the data corresponding to the maximum cost - performance ratio coefficient. Then there is: Table 1 Partial implementation data and cost - performance ratio coefficient of maximum income

[0029] According to the data in Table 1 above, it can be known that when the total minimum resource consumption invested in the 5th week is 1 (ten thousand yuan per hectare), its cost - performance ratio coefficient for obtaining the maximum income is the largest, which is 41.5% (as Figure 5 shown). Then an additional cost of 1 (ten thousand yuan per hectare) can be invested in the next 5 weeks as the total cost of fertilization, pesticide spraying, and irrigation.

[0030] And, the theoretical additional income that can be obtained by crops in the next 3 - 7 weeks is a calculation formula obtained through fitting.

[0031] Among them, the theoretical additional income that can be obtained by crops in the next 3 - 7 weeks , which can be expressed as the average additional yield of the same crop planted on the current date in history. The minimum additional resource consumption required due to environmental changes in the next 3-7 weeks It is also expressed as the resource consumption of planting the same crop on the current date in history. With minimal resource consumption Build a mathematical model of the relationship between Figure 2 As shown in the figure, the red dots are the distribution of 100 historical sample data), then: .

[0032] Example 4 like Figure 1 As shown in the figure, cross-regional collaborative multi-platform analysis of the resource redundancy of each farm is carried out, and the resource demanders and suppliers are matched through the price priority strategy; the price priority strategy is to rank the bids of the resource demanders and give priority to the delivery of crops to the dealers with the highest bids. Resource redundancy analysis refers to sharing resource data (such as surplus fertilizers and idle equipment) through multiple platforms (such as farm management systems and supply chain platforms). And a distributed database (such as MongoDB sharding cluster) is used to count the resource redundancy of each region in real time. The resource redundancy is the underutilized resources (such as surplus fertilizers and idle irrigation equipment) in a certain area.

[0033] Example 5 like Figure 1 , 6 As shown in , 7, the execution terminal adjusts the parameters of the agricultural planting equipment according to the optimization plan, and feeds back the operation results to the blockchain network. Among them, the intelligent agricultural planting management method based on AI analysis also includes: 1) Obtain images of plant growth status based on drone inspection images; 2) Extract plant growth status characteristics based on image segmentation; 3) Directly adjust the parameters of agricultural planting equipment according to the growth status characteristics of the plant (the growth status characteristics of the plant include leaf area index and pest and disease spots).

[0034] Leaf area index is the total area of ​​plant leaves per unit surface area, reflecting the density of crop growth. Pest and disease spots are plant disease areas (such as yellowing and browning) detected through image recognition.

[0035] Among them, the leaf area index of plant growth is to count the number of pixels in the leaf area and calculate the total leaf area per unit surface area by combining the actual ground area (through image resolution and scale conversion): ; In the formula, represents the leaf area index, with the unit of ㎡ / ㎡ (as Figure 6 shown); represents the total leaf area; represents the ground surface area. In mature crops, when the leaf area index it indicates that there is an abnormality in the growth state of the crops in the farmland, and the parameters of the agricultural planting equipment need to be adjusted.

[0036] Among them, the pest and disease spots of plant growth are detected for abnormal areas (such as yellowing and browning) through the HSV / YCbCr color space, and there are: ; In the formula, represents the disease density (as Figure 7 shown); represents the total area of the spots on the leaves; represents the total leaf area. In mature crops, when it indicates that there are obvious insect pests in the farmland.

[0037] Example 6 As Figure 1 shown, an intelligent agricultural planting management system based on AI analysis includes: a data collection module, which monitors soil temperature and humidity, and light intensity according to the deployed environmental sensors; obtains the real-time price and demand of crops according to the market data interface; a drone inspection unit, which obtains the plant growth state characteristics according to the drone inspection images; an interference revenue model, which generates a planting plan and a resource allocation plan based on environmental data and market dynamics; an execution control terminal, which is used to connect the irrigation system, the intelligent greenhouse, and the drone equipment, and automatically adjusts the operation parameters according to the policy instructions.

[0038] The above embodiments only represent several implementation manners of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the patent of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several modifications and improvements can still be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention should be subject to the appended claims.

Claims

1. An intelligent agricultural planting management method based on AI analysis, characterized in that, It includes the following steps: S1. Collect environmental data and market dynamics in real time, and perform data cleaning and data mapping output through edge nodes; S2. Upload the processed data to the blockchain network, generate a timestamp and encrypt and store it; S3. The interference revenue model calls blockchain data to predict the additional maintenance costs required for crop demand and environmental changes in the next 3-7 weeks, and generates an optimization plan that takes into account the maximum revenue and minimum resource consumption; S4. Cross-regionally collaborate with multiple platforms to analyze the resource redundancy of each farm, and match resource demanders and suppliers through the price priority strategy; S5. The execution terminal adjusts the parameters of agricultural planting equipment according to this optimization plan, and at the same time feeds back the operation results to the blockchain network.

2. The intelligent agricultural planting management method based on AI analysis according to claim 1, characterized in that: The intelligent agricultural planting management method based on AI analysis further includes: 1). Obtain images of the plant growth status based on drone inspection images; 2). Extract plant growth status features according to the image division; 3). Directly adjust the parameters of agricultural planting equipment according to the plant growth status features.

3. The intelligent agricultural planting management method based on AI analysis according to claim 2, characterized in that: The plant growth status features include leaf area index and pest and disease spots.

4. The intelligent agricultural planting management method based on AI analysis according to claim 1, characterized in that: The environmental data is the soil temperature and humidity, and the soil nitrogen, phosphorus and potassium content obtained based on environmental sensors; The market dynamics are the real-time prices and demand quantities of crops obtained based on market data interfaces.

5. The intelligent agricultural planting management method based on AI analysis according to claim 4, characterized in that: The data cleaning is to eliminate environmental sensor noise through Kalman filtering; The data mapping output is to calculate the additional maintenance costs required due to environmental changes in the next 3-7 weeks according to the soil temperature and humidity and the soil nitrogen, phosphorus and potassium content.

6. The intelligent agricultural planting management method based on AI analysis according to claim 1, characterized in that: The interference revenue model calculates the cost performance coefficient for obtaining the maximum revenue under the condition of minimum resource consumption through actual revenue, theoretical revenue and minimum resource consumption. The calculation formula is: ； In the formula, represents the cost performance coefficient for obtaining the maximum benefit under the condition of minimum resource consumption; represents the actual additional benefit that can be obtained from crops in the next 3-7 weeks, in ten thousand yuan per hectare; represents the theoretical additional benefit that can be obtained from crops in the next 3-7 weeks, in ten thousand yuan per hectare; and there is: ； In the formula, represents the minimum additional resource consumption required due to environmental changes in the next 3 - 7 weeks, in units of 10,000 yuan per hectare.

7. The intelligent agricultural planting management method based on AI analysis according to claim 6, characterized in that: The interference revenue model calculates the cost performance coefficient under this data according to the collected data, and ranks them according to the size of the cost performance coefficient, and adjusts the parameters of agricultural planting equipment for the data corresponding to the maximum cost performance coefficient.

8. The intelligent agricultural planting management method based on AI analysis according to claim 1, characterized in that: The price priority strategy means ranking by the bids of resource demanders and preferentially delivering crops to the dealer with the highest bid.

9. An intelligent agricultural planting management system based on AI analysis, characterized in that, Applied to the method described in any one of claims 1-8, the intelligent agricultural planting management system based on AI analysis includes: A data collection module that monitors soil temperature and humidity and light intensity according to deployed environmental sensors; and obtains the real-time prices and demand quantities of crops according to market data interfaces; A drone inspection unit that obtains plant growth status features according to drone inspection images; An interference revenue model that generates a planting plan and a resource allocation plan based on environmental data and market dynamics; An execution control terminal for connecting an irrigation system, an intelligent greenhouse, and drone equipment, and automatically adjusting operation parameters according to policy instructions.

Images