Farmland monitoring method based on environment prediction model

Abstract

The invention relates to the technical field of informatization agriculture, in particular to a farmland monitoring method based on an environment prediction model, and the method comprises the following steps: a data collection step: collecting environment data and crop planting data of a farmland; an environment prediction step: obtaining historical data of the environment data and the crop planting data, and constructing an environment prediction model; predicting the change trend of the farmland environment in the future set time; a monitoring and early warning step: performing monitoring and early warning according to the change trend, and generating an early warning prompt when an early warning condition is met; and a disaster prevention and early warning step: when an early warning prompt is received, performing accurate delimiting and image-text simulation on the farmland disaster influence area according to the early warning prompt content, the environment data acquired in real time and the crop planting number, and generating a visual disaster prevention and early warning graph. By monitoring the farmland in real time, the condition of the farmland is analyzed and mastered more accurately, timely and visually, so that corresponding management measures are taken timely, and the production efficiency and yield of the farmland are improved.

Description

Technical Field

[0001] The present invention relates to the field of information-based agricultural technology, and in particular to a farmland monitoring method based on an environmental prediction model. Background Art

[0002] The construction of high-standard farmland is crucial to national food security, agricultural product quality, and ecological safety. Long-term monitoring of the farmland environment is crucial for modernizing agricultural production, improving farmland management efficiency, reducing agricultural production risks, and promoting sustainable agricultural development. Traditional farmland monitoring methods, which primarily rely on manual observation and simple sensor data, suffer from the following limitations: Incomplete data acquisition: Traditional methods struggle to collect large-scale, multi-dimensional farmland environmental data, resulting in partial and inaccurate monitoring results. Inadequate predictive capabilities: The lack of effective environmental prediction models makes it impossible to predict farmland environmental changes in advance, making it difficult to support precision agriculture decision-making. Low intelligence: Traditional methods rely on manual experience, making it difficult to achieve automated and intelligent farmland management. Summary of the Invention

[0003] The purpose of the present invention is to provide a farmland monitoring method based on an environmental prediction model. By real-time monitoring of environmental factors such as soil and weather in farmland, the condition of farmland can be analyzed and grasped more accurately, timely and visually, so that corresponding management measures can be taken in a timely manner, thereby improving farmland production efficiency and output.

[0004] In order to achieve the above object, a farmland monitoring method based on an environmental prediction model is provided, comprising the following steps:

[0005] Data collection step: collecting farmland environmental data and crop planting data; the environmental data includes meteorological data and soil data, the soil data includes soil temperature and humidity, pH value, nutrient content and soil characteristics, as well as geographic location data; the crop planting data includes crop planting type and crop growth data; the meteorological data includes air temperature and humidity, wind speed and rainfall;

[0006] Environmental prediction step: Obtain historical data of environmental data and crop planting data, and build an environmental prediction model based on the historical data; obtain real-time data of farmland environmental data and crop planting data, and predict the changing trend of the farmland environment within a set time in the future based on the real-time data. The changing trend includes the changing trend of soil temperature and humidity, the types of crop pests and diseases, and the probability of occurrence of pests and diseases;

[0007] Monitoring and early warning steps: Monitor and warn of soil temperature and humidity change trends. When the soil temperature and humidity change trends meet the warning conditions corresponding to floods or droughts, generate a warning prompt; monitor and warn of the types of crop pests and diseases and the probability of their occurrence. When the warning conditions for crop pests and diseases that pose a high risk to crops appear, generate a warning prompt; when the probability of pests and diseases exceeds the warning threshold, generate a warning prompt;

[0008] Disaster prevention and warning steps: When receiving a warning prompt, the area affected by the farmland disaster is accurately delineated and simulated with graphics based on the warning prompt content, real-time collected environmental data and the number of crop plantings, and a visual disaster prevention and warning map is generated; the visual disaster prevention and warning map is visually displayed in the form of a chart or map.

[0009] Furthermore, in the data collection step, the collection of the environmental data and crop planting data specifically includes the following steps:

[0010] Region division sub-step: Divide the target farmland into N farmland sub-regions and mark them;

[0011] Soil data acquisition sub-step: Install sensors in each farmland sub-area to collect soil temperature, humidity, and pH values in the target farmland; set up a positioning module to collect geographic location data; organize the geographic location data, soil temperature, humidity, and pH values of multiple farmland sub-areas into soil data through a data collector and upload it;

[0012] Crop planting data acquisition sub-step: Use multispectral drones to scan N farmland sub-areas separately, analyze the nitrogen content, chlorophyll content and moisture content of related parts of crops based on multispectral data, further analyze and obtain crop growth data, and upload the crop growth data.

[0013] Furthermore, the method further comprises the following steps:

[0014] Data preprocessing step: preprocess the environmental data and crop planting data collected from historical data to ensure data quality and consistency; the preprocessing includes at least one of data cleaning, data standardization, and data dimensionality reduction.

[0015] Furthermore, the method further comprises the following steps:

[0016] Environmental prediction model construction optimization steps: Based on the preprocessed historical data, the environmental prediction model is trained using a machine learning algorithm, and the machine learning algorithm includes at least one of linear regression, support vector machine, neural network, and random forest; the trained environmental prediction model is evaluated using a cross-validation method, and the model parameters are optimized according to the evaluation results.

[0017] Furthermore, the method further comprises the following steps:

[0018] Soil moisture analysis steps: Based on the soil temperature and humidity data of each farmland sub-region, analyze the soil moisture information, and combine it with the crop growth data to assess the degree of crop water shortage and the water shortage area;

[0019] Cost analysis step: Determine the farmland water replenishment method based on the water shortage level and water shortage area, and further analyze the implementation cost of different farmland water replenishment methods based on the farmland water replenishment method, water shortage level and water shortage area; the farmland water replenishment method includes manual sprinkling, drip irrigation, sprinkler irrigation, and furrow irrigation.

[0020] Furthermore, the soil moisture analysis step also includes: based on soil moisture information and combined with meteorological data, constructing a soil moisture dynamic change model to predict the soil moisture change trend of each farmland sub-area within a preset time period in the future; according to the soil moisture change trend, dynamically adjusting the water shortage level and water shortage area of crops.

[0021] Furthermore, the cost analysis step further includes:

[0022] Collect the implementation costs and crop returns after irrigation in water-scarce areas, and establish a cost-benefit model for different farmland water replenishment methods to evaluate the economic benefits of each water replenishment method;

[0023] Based on the cost-benefit model, combined with real-time environmental data and forecast data, the optimal farmland water replenishment plan is recommended, and a corresponding cost budget and implementation plan are generated.

[0024] Furthermore, the method further comprises the following steps:

[0025] Regional cluster analysis steps: Based on the water shortage levels and water shortage areas of multiple farmland sub-regions, a clustering algorithm is used to automatically select and group adjacent or similar water shortage areas based on spatial distribution characteristics and similarity of water shortage levels;

[0026] Based on the geographical location of water-scarce areas within each farmland sub-region, spatially adjacent water-scarce areas are prioritized and grouped together. Based on the water scarcity level of each farmland sub-region, water-scarce areas with similar water scarcity levels are grouped together. Based on the soil characteristics, crop types, and meteorological data of each farmland sub-region, the grouping results are tailored to the actual needs of agricultural management.

[0027] Through cluster analysis, several water-scarce area groups were generated. The water-scarce areas in each group have similar water-scarce characteristics and spatial correlations, providing a basis for subsequent optimized allocation of water resources and precise irrigation.

[0028] Furthermore, the method further comprises the following steps:

[0029] Dynamic clustering optimization step: Based on real-time updated farmland environmental data and the changing trends predicted by the environmental prediction model, the clustering algorithm parameters and grouping results are dynamically adjusted to ensure that the division of water-scarce area groups always conforms to the current farmland environmental changes;

[0030] The dynamic clustering optimization step introduces time series analysis to identify the periodic change patterns of farmland environmental data and combines it with a machine learning model to predict the changing trends of future clustering groups.

[0031] Furthermore, the method further comprises the following steps:

[0032] Intelligent irrigation zoning control step: Based on the division results of the water-deficient area groups, an independent irrigation control strategy is generated for each area group; the irrigation control strategy includes the optimized configuration of irrigation time, irrigation amount and irrigation method, and precise control and automatic execution of each area group are achieved through Internet of Things devices.

[0033] Beneficial effects of this patent:

[0034] 1. Improve the comprehensiveness and real-time nature of farmland monitoring: By integrating multiple data collection devices, including weather stations, soil sensors, and multispectral drones, comprehensive, real-time collection of farmland environmental data is achieved. This addresses the incomplete and untimely data acquisition issues associated with traditional methods, providing reliable data support for precision agriculture decision-making.

[0035] 2. Enhanced accuracy and foresight of environmental forecasts: Environmental forecasting models based on machine learning algorithms can accurately predict future trends in farmland environments, such as soil moisture, temperature, types of crop pests and diseases, and the probability of their occurrence. This provides greater precision and depth for farmland management. Furthermore, combined with farmland monitoring, this not only improves the scientific and intelligent nature of farmland management but also provides timely and accurate decision-making support for farmers and managers in complex and changing environmental conditions. It also enables precise allocation and efficient utilization of agricultural resources. For example, based on environmental forecasts and crop needs, scientific irrigation and fertilization operation plans can be generated, avoiding resource waste.

[0036] 3. Establish a disaster prevention system to reduce agricultural production risks: Utilizing environmental forecasts and real-time monitoring of farmland environmental data and crop planting data, accurate environmental forecasting and intelligent early warning mechanisms can identify potential agricultural risks (such as droughts and floods) in advance and enable timely response measures. This reduces crop losses caused by environmental changes and ensures the stability of agricultural production. Furthermore, it can accurately delineate and simulate disaster-affected areas, generating visual disaster prevention and early warning maps to facilitate decision-making support for managers.

[0037] 4. Accurately assess crop water shortages: By analyzing soil temperature and humidity data for each farmland sub-region and combining it with crop growth data, the severity and location of crop water shortages can be accurately assessed, providing a scientific basis for irrigation decisions. Improving water resource utilization efficiency: Analysis based on soil moisture information can avoid the water waste caused by blind irrigation in traditional irrigation methods, achieving precise allocation and efficient use of water resources. Supporting healthy crop growth: By monitoring soil moisture changes in real time, crop water shortages can be promptly identified and appropriate measures can be taken to ensure healthy crop growth and improve crop yield and quality. Optimizing irrigation costs: By analyzing the costs of different water replenishment methods based on the severity and location of water shortages, farmers can select the most economical irrigation plan and reduce agricultural production costs. Improving scientific decision-making: Through cost-benefit analysis, the system comprehensively considers irrigation effectiveness and economic investment, providing farmers with scientific support for irrigation decision-making and avoiding resource waste and economic losses caused by poor decision-making. Promoting sustainable agricultural development: By optimizing irrigation costs, water waste is reduced, negative environmental impacts are minimized, and sustainable agricultural development is promoted.

[0038] The integration of soil moisture analysis and cost analysis enables the farmland monitoring system to not only provide accurate environmental data but also offer farmers practical irrigation recommendations, improving system practicality and user satisfaction. This system also enhances its intelligence by incorporating data analysis and cost optimization algorithms to automatically generate scientific irrigation plans, reducing manual intervention and improving system intelligence. Support for large-scale farmland management: Through zoning analysis and cost optimization, the system adapts to the management needs of large-scale farmland, demonstrating strong scalability and application prospects.

[0039] 5. Support for regional collaborative optimization and large-scale application: Through regional cluster analysis and collaborative optimization models, balanced allocation and efficient utilization of water resources across regions are achieved. The use of distributed computing and dynamic cluster optimization technology enables the system to adapt to large-scale farmland monitoring scenarios and has strong scalability.

[0040] By integrating advanced data acquisition technology, machine learning algorithms, Internet of Things technology and intelligent decision-making models, this invention realizes comprehensive, precise and intelligent farmland monitoring, effectively solves many problems existing in traditional farmland monitoring methods, provides strong technical support for the development of modern agriculture, and has significant social, economic and ecological benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a flowchart of a farmland monitoring method based on an environmental prediction model according to an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following is further described in detail through specific implementation methods:

[0043] Example

[0044] A farmland monitoring method based on an environmental prediction model is basically as follows Figure 1 As shown, the following steps are included:

[0045] Data collection step: Collect environmental data and crop planting data of farmland; the environmental data includes meteorological data and soil data, and the soil data includes soil temperature and humidity, pH value, nutrient content and soil characteristics, as well as geographic location data; the crop planting data includes crop planting type and crop growth data; the meteorological data includes air temperature and humidity, wind speed and rainfall; the meteorological data is obtained through a weather station. In this embodiment, the crop planting data of dry fields is mainly collected. The crop planting data and soil data are obtained by the following steps.

[0046] In the data collection step, the collection of the environmental data and crop planting data specifically includes the following steps:

[0047] Region division sub-step: Divide the target farmland into N farmland sub-regions and mark them; the overall array distribution is presented.

[0048] Soil data acquisition sub-steps: Install sensors in each farmland sub-area to collect soil temperature, humidity, and pH values in the target farmland; set up a positioning module to collect geographic location data; organize the geographic location data of multiple farmland sub-areas and soil temperature, humidity, nutrient content, soil properties (red soil, black soil, sandy soil, etc.) and pH values into soil data through a data collector and upload them; in this embodiment, a data acquisition network is constructed using Internet of Things technology, which can monitor soil data in various areas of the farmland in real time and accurately, and can ensure data acquisition efficiency and accuracy; the Internet of Things technology includes interconnection and data exchange between sensors and actuators, and data analysis. Considering the subsequent visual graphic analysis, the acquisition of geographic location data is also relatively important, and it can be collected through a conventional positioning module.

[0049] Crop planting data acquisition sub-step: Use multispectral drones to scan N farmland sub-areas separately, analyze the nitrogen content, chlorophyll content and moisture content of related parts of crops based on multispectral data, further analyze and obtain crop growth data, and upload the crop growth data.

[0050] Data preprocessing: Collected historical environmental and crop data is preprocessed to ensure data quality and consistency. This preprocessing includes at least one of data cleaning, data standardization, and data dimensionality reduction. Data cleaning: Addresses missing values and outliers. Data standardization: Converts data of different dimensions to the same scale, such as normalizing to the [0, 1] interval. Data dimensionality reduction: For high-dimensional data, methods such as principal component analysis (PCA) can be used to reduce dimensionality and computational complexity.

[0051] Environmental prediction steps: obtain historical data of environmental data and crop planting data, and build an environmental prediction model based on the historical data; obtain real-time data of farmland environmental data and crop planting data, and predict the changing trend of the farmland environment within a set time in the future based on the real-time data. The changing trend includes the changing trend of soil temperature and humidity, the types of crop diseases and pests, and the probability of occurrence of diseases and pests.

[0052] Environmental prediction model construction and optimization steps: Based on preprocessed historical data, the environmental prediction model is trained using a machine learning algorithm, including at least one of linear regression, support vector machine, neural network, and random forest. The trained environmental prediction model is evaluated using cross-validation, and the model parameters are optimized based on the evaluation results. Linear regression: Suitable for scenarios with obvious linear relationships, the model is simple and easy to interpret. Support vector machine (SVM): Suitable for high-dimensional data and capable of handling nonlinear relationships. Neural network: Capable of learning complex nonlinear relationships, it has high prediction accuracy, but the model has poor interpretability. Random forest: Composed of multiple decision trees, it can handle high-dimensional data and has good generalization capabilities. The selection can be based on the farm size and customer needs. For model training, divide the preprocessed data into training and test sets: Divide the preprocessed data into training and test sets, for example, in a 7:3 ratio. Model training: Use the training set data to train the selected machine learning model, adjusting the model parameters to ensure that the model fits the data well. Model evaluation: Use the test set data to evaluate the prediction performance of the model. Common evaluation indicators include mean square error (MSE), mean absolute error (MAE), coefficient of determination (R 2 ) etc. Model Optimization: Based on the model evaluation results, the model parameters are adjusted and optimized, for example, using grid search, random search, and other methods to find the optimal parameter combination. As farmland environments change over time, the environmental prediction model needs to be regularly updated to ensure its prediction accuracy. The model can be retrained with new data or updated using online learning algorithms. The construction of an environmental prediction model is an iterative optimization process, requiring continuous adjustment of model parameters and algorithms based on actual conditions to improve the model's prediction accuracy and generalization capabilities.

[0053] Monitoring and early warning steps: Monitor and warn of the changing trends of soil temperature and humidity. When the changing trends of soil temperature and humidity meet the warning conditions corresponding to floods or droughts, generate early warning prompts; monitor and warn of the types of crop diseases and pests and the probability of their occurrence. When the warning conditions for high-risk crop diseases and pests to crops appear, generate early warning prompts; when the probability of disease and pest occurrence exceeds the warning threshold, generate early warning prompts; by monitoring and predicting the farmland environment, managers can promptly discover and respond to various natural disasters (such as droughts, floods, diseases and pests, etc.), and take corresponding measures to mitigate the impact of disasters on farmland and reduce losses.

[0054] Disaster prevention and warning steps: When receiving a warning prompt, the area affected by the farmland disaster is accurately delineated and simulated with graphics based on the warning prompt content, real-time collected environmental data and the number of crop plantings, and a visual disaster prevention and warning map is generated; the visual disaster prevention and warning map is visually displayed in the form of a chart or map.

[0055] This solution integrates multiple data collection devices, including weather stations, soil sensors, and multispectral drones, to achieve comprehensive, real-time collection of farmland environmental data. This overcomes the incomplete and untimely data acquisition issues associated with traditional methods, providing reliable data support for precision agriculture decision-making. An environmental prediction model, built based on machine learning algorithms, accurately predicts future trends in farmland environmental changes, such as soil moisture, temperature, types of crop pests and diseases, and the probability of their occurrence. This provides greater precision and depth for farmland management. Furthermore, combined with farmland monitoring, this not only enhances the scientific and intelligent nature of farmland management but also provides timely and accurate decision-making support for farmers and managers in complex and changing environmental conditions. It also enables precise allocation and efficient utilization of agricultural resources. For example, based on environmental predictions and crop needs, it generates scientific irrigation and fertilization operation plans, avoiding resource waste.

[0056] Soil moisture analysis step: Based on the soil temperature and humidity data of each farmland sub-area, the soil moisture information is analyzed, and combined with the crop growth data, the water shortage level of the crops and the water shortage area are evaluated; the soil moisture analysis step also includes: based on the soil moisture information, combined with meteorological data, a soil moisture dynamic change model is constructed to predict the soil moisture change trend of each farmland sub-area within a preset time period in the future; according to the soil moisture change trend, the water shortage level of the crops and the water shortage area are dynamically adjusted.

[0057] This solution can accurately assess the water shortage of crops through the soil moisture analysis step. By analyzing the soil temperature and humidity data of each farmland sub-area and combining it with crop growth data, it can accurately assess the degree of water shortage of crops and the water shortage area, providing a scientific basis for irrigation decision-making. Improve the efficiency of water resource utilization: Based on the analysis of soil moisture information, it can avoid the waste of water resources caused by blind irrigation in traditional irrigation methods and achieve accurate allocation and efficient utilization of water resources. At the same time, it supports the healthy growth of crops. For example, by real-time monitoring of soil moisture changes, it can promptly detect crop water shortage problems and take corresponding measures to ensure the healthy growth of crops and improve crop yield and quality.

[0058] Cost analysis: Determine the farmland water replenishment method based on the water shortage level and water shortage area, and further analyze the implementation costs of different farmland water replenishment methods based on the farmland water replenishment method, water shortage level, and water shortage area. Farmland water replenishment methods include manual sprinkling, drip irrigation, sprinkler irrigation, and furrow irrigation. The cost analysis step also includes collecting the implementation costs and crop returns after irrigation in the water-scarce areas, establishing a cost-benefit model for different farmland water replenishment methods, and evaluating the economic benefits of each water replenishment method. Based on the cost-benefit model, combined with real-time environmental data and forecast data, recommend the optimal farmland water replenishment plan and generate a corresponding cost budget and implementation plan.

[0059] This program optimizes irrigation costs through cost analysis. For example, it analyzes the costs of different water replenishment methods for farmland based on the degree of water shortage and the area of water shortage, helping farmers select the most economical irrigation plan and reduce agricultural production costs. It also improves the scientific nature of decision-making: Through cost-benefit analysis, it comprehensively considers irrigation effectiveness and economic investment, providing farmers with scientific irrigation decision-making support and avoiding resource waste and economic losses caused by inappropriate decisions. It also promotes sustainable agricultural development: By optimizing irrigation costs, it reduces water waste, mitigates negative environmental impacts, and promotes sustainable agricultural development.

[0060] Regional clustering analysis steps: for the water shortage levels and water shortage areas of multiple farmland sub-regions, based on the spatial distribution characteristics and similarity of water shortage levels, clustering algorithms (such as K-means, DBSCAN or hierarchical clustering algorithms) are used to automatically select and group adjacent or similar water shortage areas; according to the geographical location information of the water shortage areas in each farmland sub-region, spatially adjacent water shortage areas are preferentially divided into the same group; according to the water shortage level of the water shortage areas in each farmland sub-region, water shortage areas with similar water shortage levels are divided into the same group; according to the soil characteristics, crop types and meteorological data of each farmland sub-region, the grouping results are made to meet the actual needs of agricultural management; through cluster analysis, several water shortage area groups are generated, and the water shortage areas in each group have similar water shortage characteristics and spatial correlation, providing a basis for subsequent water resource optimization allocation and precision irrigation.

[0061] This solution uses a clustering algorithm to automatically group regions, eliminating the subjectivity and inefficiency of manual division. It supports multiple clustering algorithms, allowing users to select the most appropriate method based on specific scenarios. It also comprehensively considers multi-dimensional features, including not only spatial location and water scarcity, but also soil characteristics, crop type, and meteorological conditions. This makes the grouping results more scientific and practical, providing a foundation for subsequent optimization and supporting optimal water resource allocation and precision irrigation.

[0062] Dynamic clustering optimization: Based on real-time farmland environmental data and the changing trends predicted by the environmental prediction model, the clustering algorithm parameters and grouping results are dynamically adjusted to ensure that the water-scarce area groupings remain consistent with current farmland environmental changes. This dynamic clustering optimization step uses time series analysis to identify the cyclical changes in farmland environmental data and combines it with a machine learning model to predict future clustering trends. This dynamic adjustment mechanism enables clustering groups to be updated in real time as the environment changes, improving the overall adaptability and accuracy of the solution. Combining time series analysis with machine learning predictions enhances the system's ability to predict changes in the farmland environment.

[0063] Intelligent irrigation zoning control steps: Based on the water-deficient area grouping results, an independent irrigation control strategy is generated for each area group. This irrigation control strategy includes optimized configuration of irrigation time, irrigation amount, and irrigation method. IoT devices are used to achieve precise control and automated execution of each area group. This solution, based on clustering results, implements zoned precision irrigation, avoiding the resource waste associated with traditional irrigation methods. This allows for automated control through IoT technology, improving irrigation efficiency and management.

[0064] This solution combines soil moisture analysis, cost analysis, dynamic clustering optimization, and intelligent irrigation zoning control. This enables the farmland monitoring system to not only provide accurate environmental data but also offer farmers practical irrigation recommendations, improving system practicality and user satisfaction. It also enhances the system's intelligence by incorporating data analysis and cost optimization algorithms to automatically generate scientific irrigation plans, reducing manual intervention and improving the system's intelligence. It also supports large-scale farmland management: Through zoning analysis and cost optimization, the system adapts to the management needs of large-scale farmland, demonstrating strong scalability and application prospects.

[0065] Visual display step: Visually display the visual disaster prevention warning map generated in the disaster prevention warning step through a display device, and visually display the irrigation control strategy of the intelligent irrigation zoning control step through a display device.

[0066] The above is only an embodiment of the present invention. Common knowledge such as the known specific structures and characteristics in the scheme will not be described in detail here. Those of ordinary skill in the art are aware of all common technical knowledge in the technical field to which the invention belongs before the application date or priority date, can obtain all existing technologies in the field, and have the ability to apply conventional experimental means before that date. Those of ordinary skill in the art can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for those of ordinary skill in the art to implement this application. It should be pointed out that for those skilled in the art, several variations and improvements can be made without departing from the structure of the present invention. These should also be regarded as the scope of protection of the present invention. These will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A farmland monitoring method based on an environmental prediction model, characterized in that: The following steps are involved: Data collection step: collecting farmland environmental data and crop planting data; the environmental data includes meteorological data and soil data, the soil data includes soil temperature and humidity, pH value, nutrient content and soil characteristics, as well as geographic location data; the crop planting data includes crop planting type and crop growth data; the meteorological data includes air temperature and humidity, wind speed and rainfall; Environmental prediction step: Obtain historical data of environmental data and crop planting data, and build an environmental prediction model based on the historical data; obtain real-time data of farmland environmental data and crop planting data, and predict the changing trend of the farmland environment within a set time in the future based on the real-time data. The changing trend includes the changing trend of soil temperature and humidity, the types of crop pests and diseases, and the probability of occurrence of pests and diseases; Monitoring and early warning steps: Monitor and warn of soil temperature and humidity change trends. When the soil temperature and humidity change trends meet the warning conditions corresponding to floods or droughts, generate early warning prompts; Monitor and warn of the types of crop pests and diseases and the probability of their occurrence, and generate warning alerts when warning conditions for crop pests and diseases that pose a high risk to crops appear; When the probability of pests and diseases exceeding the warning threshold, an early warning prompt is generated; Disaster prevention and early warning steps: When receiving an early warning, accurately delineate and simulate the farmland disaster-affected area based on the warning content, real-time environmental data, and crop planting numbers, and generate a visual disaster prevention and early warning map; The visual disaster prevention and warning map is visually displayed in the form of a chart or a map.

2. The farmland monitoring method based on the environmental prediction model according to claim 1, characterized in that: In the data collection step, the collection of the environmental data and crop planting data specifically includes the following steps: Region division sub-step: Divide the target farmland into N farmland sub-regions and mark them; Soil data acquisition sub-step: Install sensors in each farmland sub-area to collect soil temperature, humidity, and pH values in the target farmland; set up a positioning module to collect geographic location data; organize the geographic location data, soil temperature, humidity, and pH values of multiple farmland sub-areas into soil data through a data collector and upload it; Crop planting data acquisition sub-step: Use multispectral drones to scan N farmland sub-areas separately, analyze the nitrogen content, chlorophyll content and moisture content of related parts of crops based on multispectral data, further analyze and obtain crop growth data, and upload the crop growth data.

3. The farmland monitoring method based on the environmental prediction model according to claim 2, characterized in that: The following steps are also included: Data preprocessing step: Preprocess the collected historical environmental data and crop planting data to ensure data quality and consistency; The preprocessing includes at least one of data cleaning, data standardization, and data dimensionality reduction.

4. The farmland monitoring method based on the environmental prediction model according to claim 3, characterized in that: The following steps are also included: Environmental prediction model construction optimization steps: Based on the preprocessed historical data, the environmental prediction model is trained using a machine learning algorithm, and the machine learning algorithm includes at least one of linear regression, support vector machine, neural network, and random forest; the trained environmental prediction model is evaluated using a cross-validation method, and the model parameters are optimized according to the evaluation results.

5. The farmland monitoring method based on the environmental prediction model according to claim 4, characterized in that: The following steps are also included: Soil moisture analysis steps: Based on the soil temperature and humidity data of each farmland sub-region, analyze the soil moisture information, and combine it with the crop growth data to assess the degree of crop water shortage and the water shortage area; Cost analysis step: Determine the farmland water replenishment method based on the water shortage level and water shortage area, and further analyze the implementation cost of different farmland water replenishment methods based on the farmland water replenishment method, water shortage level and water shortage area; the farmland water replenishment method includes manual sprinkling, drip irrigation, sprinkler irrigation, and furrow irrigation.

6. The farmland monitoring method based on the environmental prediction model according to claim 5, characterized in that: The soil moisture analysis step also includes: constructing a soil moisture dynamic change model based on soil moisture information and combined with meteorological data to predict the soil moisture change trend of each farmland sub-area within a preset time period in the future; and dynamically adjusting the water shortage level and water shortage area of crops according to the soil moisture change trend.

7. The farmland monitoring method based on the environmental prediction model according to claim 6, characterized in that: The cost analysis step further includes: Collect the implementation costs and crop returns after irrigation in water-scarce areas, and establish a cost-benefit model for different farmland water replenishment methods to evaluate the economic benefits of each water replenishment method; Based on the cost-benefit model, combined with real-time environmental data and forecast data, the optimal farmland water replenishment plan is recommended, and a corresponding cost budget and implementation plan are generated.

8. The farmland monitoring method based on the environmental prediction model according to claim 7, characterized in that: The following steps are also included: Regional cluster analysis steps: Based on the water shortage levels and water shortage areas of multiple farmland sub-regions, a clustering algorithm is used to automatically select and group adjacent or similar water shortage areas based on spatial distribution characteristics and similarity of water shortage levels; Based on the geographical location of water-scarce areas within each farmland sub-region, spatially adjacent water-scarce areas are prioritized and grouped together. Based on the water scarcity level of each farmland sub-region, water-scarce areas with similar water scarcity levels are grouped together. Based on the soil characteristics, crop types, and meteorological data of each farmland sub-region, the grouping results are tailored to the actual needs of agricultural management. Through cluster analysis, several water-scarce area groups were generated. The water-scarce areas in each group have similar water-scarce characteristics and spatial correlations, providing a basis for subsequent optimized allocation of water resources and precise irrigation.

9. The farmland monitoring method based on the environmental prediction model according to claim 8, characterized in that: The following steps are also included: Dynamic clustering optimization step: Based on real-time updated farmland environmental data and the changing trends predicted by the environmental prediction model, the clustering algorithm parameters and grouping results are dynamically adjusted to ensure that the division of water-scarce area groups always conforms to the current farmland environmental changes; The dynamic clustering optimization step introduces time series analysis to identify the periodic change patterns of farmland environmental data and combines it with a machine learning model to predict the changing trends of future clustering groups.

10. The farmland monitoring method based on the environmental prediction model according to claim 9, characterized in that: The following steps are also included: Intelligent irrigation zoning control step: generating an independent irrigation control strategy for each area group based on the division result of the water-deficient area group; The irrigation control strategy includes the optimized configuration of irrigation time, irrigation amount and irrigation method, and realizes precise control and automatic execution of each regional group through IoT devices.

Images