Farmland automatic irrigation control system

Abstract

The invention discloses an automatic farmland irrigation control system, and relates to the technical field of farmland irrigation. The farmland automatic irrigation control system comprises a time sequence state acquisition unit used for acquiring time sequence data of land, plants and environment states of each irrigation area in a farmland; the time sequence feature extraction unit is used for extracting time sequence feature data of each region on the basis of a pre-trained model; the judgment and analysis unit is used for calculating and analyzing an irrigation demand index of each area based on the extracted data and comparing the irrigation demand index with a preset demand interval; the time sequence state data acquisition unit is introduced, the land, plant and environment state data are collected and processed in real time, effective time sequence characteristics are extracted, the irrigation demand index is accurately calculated, the actual area, the soil moisture content and other parameters are combined, and the real-time monitoring of the irrigation demand index is achieved. The system can reasonably calculate the irrigation amount and duration, water resource waste is avoided, and therefore the water resource utilization efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of farmland irrigation technology, specifically to an automated farmland irrigation control system. Background Technology

[0002] With global population growth and the impact of climate change, agricultural production faces increasing challenges, especially in water-scarce regions. Precision irrigation has become crucial for improving agricultural efficiency and sustainable development. Traditional irrigation methods, often based on experience or fixed schedules, easily lead to water waste or insufficient irrigation, which not only affects crop growth but also exacerbates water shortages.

[0003] Furthermore, the limitations of existing technologies include at least the following problems: traditional irrigation control systems typically rely on static preset irrigation plans or simple soil moisture sensors for decision-making. This approach often ignores the dynamic changes in various irrigation areas within the farmland. Due to the lack of real-time and accurate time-series status data, existing systems struggle to monitor and analyze the comprehensive changes in soil, plant, and environmental conditions in real time, leading to biased judgments of irrigation needs. This can easily result in insufficient or excessive irrigation, which not only wastes precious water resources but may also negatively impact crop growth, reducing crop yield and quality and increasing agricultural production costs.

[0004] Traditional methods often struggle to flexibly adjust irrigation volumes based on real-time changes, lack intelligent adaptability, and are ill-equipped to cope with varying water resource demands under different climatic conditions. Consequently, they fail to achieve the goals of precision agriculture. This problem severely restricts the improvement of irrigation efficiency and leads to resource waste and instability in agricultural production. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an automated irrigation control system for farmland, which solves the problem that traditional irrigation control systems struggle to dynamically adjust irrigation demand, leading to water waste and insufficient irrigation.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an automated irrigation control system for farmland, comprising: a time-series state acquisition unit, which acquires time-series data of the state of several irrigation areas within the farmland, including time-series data of land, plants, and environment; a time-series feature extraction unit, which extracts features from the time-series data of the state of each irrigation area within the farmland based on a pre-trained time-series state feature extraction model, to obtain the time-series data of the state of the corresponding irrigation area, including a set of time-series features of land, plants, and environment; a judgment and analysis unit, which analyzes the irrigation demand index of each irrigation area within the farmland based on the time-series data of the state, and performs judgment and analysis with a preset irrigation demand interval; and an irrigation execution unit, which considers irrigation areas whose irrigation demand index falls within the preset irrigation demand interval as areas requiring irrigation and executes irrigation measures.

[0007] Furthermore, the land condition time series data includes soil moisture content, soil osmotic resistance, and soil temperature values ​​at several time points; the plant condition time series data includes canopy infrared temperature and leaf near-infrared reflectance values ​​at several time points; and the environmental condition time series data includes air temperature, air humidity, and micro-area evaporation offset values ​​at several time points.

[0008] Furthermore, the land state temporal feature set includes the slope value of water content change, the standard deviation value of osmotic impedance, and the amplitude value of soil temperature change; the plant state temporal feature set includes the canopy temperature difference range value, the rate of change of leaf reflectance, and the standard deviation value of leaf reflectance; the environmental state temporal feature set includes the slope value of air temperature change, the amplitude value of humidity change, and the average value of evaporation offset; and the temporal state feature extraction model includes a land state feature extraction network, a plant state feature extraction network, and an environmental state feature extraction network.

[0009] Further, the specific steps for obtaining the regional state time series data of the corresponding irrigation area are as follows: In the land state feature extraction network of the time series state feature extraction model, the land state time series data of each irrigation area in the farmland is subjected to multi-time point statistical analysis to obtain the land state time series feature set of the corresponding irrigation area; In the plant state feature extraction network of the time series state feature extraction model, the plant state time series data of each irrigation area in the farmland is subjected to multi-time point fluctuation analysis to obtain the plant state time series feature set of the corresponding irrigation area; In the environmental state feature extraction network of the time series state feature extraction model, the environmental state time series data of each irrigation area in the farmland is subjected to trend extraction and statistical processing to obtain the environmental state time series feature set of the corresponding irrigation area.

[0010] Furthermore, the specific steps for analyzing the irrigation demand index of each irrigated area within the farmland are as follows: Based on the regional state time series data, analyze the time series characteristic index set of each irrigated area within the farmland, including the land, plant, and environmental state time series characteristic indices; based on the time series characteristic index set, analyze the irrigation demand index of each irrigated area within the farmland.

[0011] Furthermore, the specific formula for calculating the irrigation demand index of a certain irrigated area within farmland is as follows: ;in, This represents the irrigation demand index for a specific irrigated area within farmland. This refers to the temporal characteristic index of land status in a specific irrigated area within farmland. This refers to the temporal characteristic index of plant status in a certain irrigated area of ​​farmland. These are the plant response coefficients stored in the database. This refers to the temporal characteristic index of the environmental state of a certain irrigated area within farmland. These are the environmental response coefficients stored in the database.

[0012] Further, the specific steps for analyzing the temporal characteristic index of land state in each irrigated area within the farmland are as follows: read the slope value of water content change, the standard deviation value of osmotic impedance, and the amplitude value of soil temperature change for each irrigated area within the farmland, and perform standardization processing; based on the standardized slope value of water content change, the standard deviation value of osmotic impedance, and the amplitude value of soil temperature change, analyze the temporal characteristic index of land state in each irrigated area within the farmland.

[0013] Further, the specific steps for analyzing the temporal characteristic index of plant status in each irrigated area of ​​the farmland are as follows: read the canopy temperature difference range and leaf reflectance change rate of each irrigated area in the farmland, and perform standardization processing; combine the standard deviation of leaf reflectance of each irrigated area in the farmland with the standardized canopy temperature difference range and leaf reflectance change rate of the corresponding irrigated area for comprehensive analysis to obtain the temporal characteristic index of plant status in each irrigated area of ​​the farmland.

[0014] Further, the specific steps for analyzing the environmental state time-series characteristic index of each irrigated area in the farmland are as follows: read the humidity change amplitude value and evaporation offset average value of each irrigated area in the farmland and perform standardization processing; combine the temperature change slope value of each irrigated area in the farmland with the standardized humidity change amplitude value and evaporation offset average value of the corresponding irrigated area for comprehensive analysis to obtain the environmental state time-series characteristic index of each irrigated area in the farmland.

[0015] Furthermore, the specific steps for implementing irrigation measures in the irrigated areas of farmland are as follows: Obtain irrigation status data of the irrigated areas of farmland, including actual area value, the difference between current soil moisture content and target soil moisture content, current soil osmotic resistance value and soil water absorption rate, and analyze irrigation water volume and irrigation duration; Based on irrigation water volume and irrigation duration, implement irrigation measures in the irrigated areas.

[0016] The present invention has the following beneficial effects: 1. This automated irrigation control system for farmland introduces a time-series status data acquisition unit to collect real-time status data of the land, plants, and environment within the farmland. This makes irrigation decisions more dynamic and precise. The time-series feature extraction unit processes this data in real time to extract useful time-series feature sets, thereby accurately calculating the irrigation demand index for each irrigation area. Combined with parameters such as actual area, soil moisture content, and osmotic resistance, the system can accurately determine the actual water demand of each area and calculate reasonable irrigation volume and duration. Through this precise control, water waste in traditional irrigation methods is avoided, water resource utilization efficiency is maximized, and crops receive sufficient water to reach optimal growth. This precise irrigation management not only improves farmland water use efficiency but also saves irrigation costs, contributing to sustainable agricultural development.

[0017] 2. This automated irrigation control system for farmland, by combining temporal data on land, plants, and the environment, can comprehensively consider the dynamic changes in climate, soil, and crops, and adjust irrigation strategies in real time. Through a temporal feature extraction network, the system analyzes environmental factors such as temperature changes and humidity fluctuations, and combines parameters such as soil permeability and water absorption rate to calculate the most suitable irrigation demand for each irrigation area. Especially in the case of large weather changes, the system can automatically judge and adjust the irrigation volume based on real-time data to avoid inappropriate irrigation timing. Through this intelligent process, the system makes farmland irrigation more automated and efficient, reduces the need for manual intervention, and improves the sustainability and intelligence level of agricultural production.

[0018] 3. This automated irrigation control system for farmland, through a time-series state feature extraction unit, can extract key features from changes in land conditions, plant growth, and environmental factors, such as the standard deviation of soil osmotic resistance, the range of plant canopy temperature difference, and the amplitude of environmental humidity changes. This allows for a comprehensive assessment of the needs of each irrigation area, enabling irrigation decisions to move beyond traditional fixed patterns and automatically adjust based on real-time data. This improves the scientific rigor and accuracy of irrigation decisions. Compared to traditional static irrigation schemes, the system can more effectively address changes in different areas of farmland, ensuring each area receives the most suitable water supply and effectively improving crop growth and water resource utilization efficiency. Through this automated decision-making process, agricultural management becomes more systematic, reducing human intervention and making irrigation management more precise and reliable.

[0019] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0020] Figure 1 This is a block diagram of an automated irrigation control system for farmland according to the present invention.

[0021] Figure 2 This is a flowchart illustrating the specific steps involved in analyzing the irrigation demand index of each irrigation area within a farmland in an automated irrigation control system of the present invention.

[0022] Figure 3 This is a sequence diagram of irrigation demand indexes for five irrigation areas within a farmland in an automated irrigation control system for farmland according to the present invention. Detailed Implementation

[0023] Please see Figure 1 This invention provides a technical solution: an automated irrigation control system for farmland, comprising: a time-series status acquisition unit, which acquires time-series data of the status of several irrigation areas within the farmland, including time-series data of land, plants, and environment; a time-series feature extraction unit, which extracts features from the time-series data of the status of each irrigation area within the farmland based on a pre-trained time-series status feature extraction model, to obtain the time-series data of the status of the corresponding irrigation area, including a set of time-series features of land, plants, and environment; a judgment and analysis unit, which analyzes the irrigation demand index of each irrigation area within the farmland based on the time-series data of the status, and performs judgment and analysis with a preset irrigation demand interval; and an irrigation execution unit, which considers irrigation areas whose irrigation demand index falls within the preset irrigation demand interval as areas requiring irrigation and executes irrigation measures.

[0024] The land condition time series data includes soil moisture content, soil osmotic resistance, and soil temperature values ​​at several time points. The plant condition time series data includes canopy infrared temperature and leaf near-infrared reflectance values ​​at several time points. The environmental condition time series data includes air temperature, air humidity, and micro-area evaporation offset values ​​at several time points.

[0025] Soil moisture content refers to the proportion of water in the soil, reflecting the degree of soil saturation. Too low a moisture content can lead to water shortage in plants, while too high a content can cause root hypoxia, affecting plant growth. To obtain this data, soil moisture sensors are used; these sensors are typically buried in the soil and determine the moisture content by measuring changes in soil moisture.

[0026] Soil osmotic impedance reflects the soil's resistance to water infiltration. A high impedance value indicates that water infiltrates the soil slowly, which may be due to denser soil particles or a higher content of organic matter. Measuring this impedance typically requires the use of an osmometer, which calculates the osmotic impedance by measuring the rate of water infiltration in the soil. This allows for the assessment of the soil's water permeability.

[0027] Soil temperature has a significant impact on plant root growth and water absorption. Both excessively high and low temperatures can negatively affect root activity. Soil temperature is typically measured using soil temperature sensors, which are buried in the soil and periodically record temperature changes.

[0028] Canopy infrared temperature refers to the measured value of plant surface temperature, usually obtained through infrared temperature sensors or remote sensing technology. Changes in canopy temperature are generally closely related to plant transpiration (the process by which plants release water) and ambient temperature. Excessively high plant temperatures may indicate that the plant is facing water stress, in which case increased irrigation is needed to help cool it down.

[0029] Near-infrared reflectance values ​​of leaves indicate a plant's ability to reflect near-infrared light. Healthy plant leaves have high reflectance, while the reflectance decreases when the plant is underwater or damaged. By using a near-infrared reflectance sensor, the reflectance values ​​of plant leaves at different times can be measured, thereby determining the plant's health status.

[0030] Air temperature affects plant growth and soil moisture evaporation. Excessively high temperatures accelerate soil moisture evaporation, thus increasing irrigation needs. By installing temperature sensors in weather stations, time-series data on air temperature can be obtained, which can be used to determine the environmental heat load and its impact on irrigation requirements.

[0031] Air humidity reflects the moisture content in the air. Higher humidity results in a lower rate of water evaporation, while lower humidity leads to a higher rate of evaporation. Humidity sensors can be used to measure air humidity regularly, and fluctuations in humidity affect the intensity of plant transpiration.

[0032] Micro-area evaporation offset refers to data on changes in soil moisture evaporation rates, typically measured using micrometeorological equipment. These devices capture differences in moisture evaporation within small areas and calculate offset values ​​to reflect changes in evaporation intensity. Evaporation offset values ​​fluctuate with changes in weather and soil conditions.

[0033] Specifically, the land state time series feature set includes the slope value of water content change, the standard deviation value of osmotic impedance, and the amplitude value of soil temperature change; the plant state time series feature set includes the canopy temperature difference range value, the rate of change of leaf reflectance, and the standard deviation of leaf reflectance; the environmental state time series feature set includes the slope value of air temperature change, the amplitude value of humidity change, and the average value of evaporation offset; and the time series state feature extraction model includes the land state feature extraction network, the plant state feature extraction network, and the environmental state feature extraction network.

[0034] The land status feature extraction network includes: The input layer (used to receive raw land condition time-series data, such as soil moisture content, soil osmotic impedance, and soil temperature).

[0035] Convolutional layers (used to extract local features from land condition data, such as the spatial distribution and temporal variation trends of soil).

[0036] Pooling layers (used to downsample the features extracted by convolutional layers, reduce data dimensionality, and retain the most important information).

[0037] Fully connected layer (used to integrate the extracted features into a final land state temporal feature set for subsequent processing).

[0038] The plant state feature extraction network includes: The input layer (used to receive raw plant status time-series data, such as canopy infrared temperature values ​​and leaf near-infrared reflectance values).

[0039] Convolutional layers (used to extract spatial and temporal features from plant state data, such as temperature fluctuations and reflectance changes).

[0040] Recurrent neural network layers (used to capture temporal features of plant states and handle time dependencies and dynamic changes in data).

[0041] Fully connected layer (used to integrate the extracted features to generate a plant state temporal feature set).

[0042] The environmental state feature extraction network includes: The input layer (used to receive raw environmental state time-series data, such as air temperature, air humidity, and micro-area evaporation offset).

[0043] Convolutional layers (used to extract local features from environmental state data, such as fluctuations in temperature and humidity).

[0044] Long Short-Term Memory (LSTM) layer (used to model long-term dependencies in environmental state data and capture trends in environmental changes).

[0045] Fully connected layer (used to integrate features extracted from environmental data to form the final environmental state time series feature set).

[0046] The pre-training steps for the temporal state feature extraction model are as follows: Large-scale historical datasets need to be collected and processed. These datasets contain time-series data on soil, plant, and environmental conditions at different points in time, including multi-dimensional information such as soil moisture, temperature, plant growth status, and air humidity. Data sources can include sensors, remote sensing technology, etc. All data undergoes standardization, i.e., de-unitization and normalization, so that data of different dimensions can be trained on the same scale, while avoiding the situation where a certain feature dominates the model learning. At this point, the data preparation work is complete.

[0047] The model is trained using collected and processed data. The training process involves inputting data into the network's convolutional layers, recurrent neural network layers (such as LSTM layers), and other layers that process time-series data to learn the temporal dependencies and feature patterns in the data. Through backpropagation and optimization algorithms (such as the Adam optimizer), the model gradually adjusts the weight parameters in the network to minimize the loss function and extract effective time-series features. Cross-validation and early stopping are used to avoid overfitting during training.

[0048] In the final stage of pre-training, the model is evaluated and validated. By testing the training set with independent validation datasets, the model's performance on different datasets is evaluated to ensure that the extracted features can effectively reflect actual soil, plant, and environmental changes. The model evaluation includes multiple indicators such as accuracy, precision, and recall to ensure that the model can effectively predict irrigation needs and generate correct irrigation decisions. If the model performance does not meet expectations, it can be further optimized by adjusting the network structure, optimizing the algorithm, or adding datasets.

[0049] The specific steps to obtain the regional state time series data of the corresponding irrigation area are as follows: In the land state feature extraction network of the time-series feature extraction model, the land state time-series data of each irrigated area in the farmland is subjected to multi-time-point statistical analysis to obtain the corresponding land state time-series feature set of the irrigated area. Specifically, the land state time-series data of each irrigated area in the farmland are received, including soil moisture content, soil osmotic impedance and soil temperature. These data are collected at multiple time points by sensors or monitoring equipment and reflect different land states. The land state feature extraction network uses convolutional layers to perform preliminary feature extraction on these time-series data. The convolutional layers will identify local features from the original data, such as soil moisture change patterns and temperature fluctuations. After the convolution operation, the data will be downsampled through pooling layers to reduce the dimensionality of the data and retain the most important feature information. After processing by fully connected layers, the extracted features will be further integrated to generate a complete land state time-series feature set. In the plant state feature extraction network of the time-series state feature extraction model, multi-time-point fluctuation analysis is performed on the plant state time-series data of each irrigated area in the farmland to obtain the plant state time-series feature set of the corresponding irrigated area. Specifically, the network receives plant state time-series data of each irrigated area in the farmland, including canopy infrared temperature values ​​and leaf near-infrared reflectance values. These data are collected at multiple time points based on sensors and remote sensing equipment in the farmland, reflecting the growth status and health status of the plants. The plant state feature extraction network first processes these data through convolutional layers to identify local features of the plant state, such as changes in canopy temperature and fluctuations in leaf reflectance. The network captures the time-series features of the plant state through recurrent neural network (RNN) layers, especially long short-term memory (LSTM) layers. This can handle the long-term dependence and dynamic changes in plant growth, especially how plants self-regulate with changes in time and environment. Through the fluctuation analysis of these time-series data, the network can identify the sensitivity of plants to environmental changes and possible abnormalities in the plant growth process. After these steps, the extracted temporal features are finally integrated through a fully connected layer to generate a plant state temporal feature set. In the environmental state feature extraction network of the time-series feature extraction model, trend extraction and statistical processing are performed on the environmental state time-series data of each irrigated area in the farmland to obtain the corresponding environmental state time-series feature set of the irrigated area. Specifically, the network receives environmental state time-series data of each irrigated area in the farmland, including air temperature, air humidity, and micro-area evaporation offset values. These data are collected by sensors at different time points, reflecting the environmental conditions of the farmland. The environmental state feature extraction network first performs preliminary feature extraction on these environmental data through convolutional layers to identify local features in the data, such as the trend of air temperature change and humidity fluctuation. Then, through the Long Short-Term Memory (LSTM) layer, the network can capture the long-term dependencies in the environmental state data, such as how temperature and humidity gradually affect soil moisture and plant growth with seasonal changes. The LSTM layer can effectively handle the trend changes in the time-series data and extract the long-term impact of environmental conditions on agricultural production. Finally, the environmental state feature extraction network integrates all the features extracted from the time-series data through a fully connected layer to generate the environmental state time-series feature set.

[0050] In this implementation plan, by collecting and extracting time-series data on the land, plants, and environmental conditions in irrigated farmland, the intelligence and precision of the irrigation system are greatly improved. By using a time-series feature extraction model, the system can capture and analyze dynamic data such as soil moisture, temperature changes, plant health status, and environmental factors in different irrigated areas in real time. This not only avoids the problem of slow response to environmental changes in traditional irrigation methods, but also enables precise irrigation decisions under different climate and soil conditions. For example, the land condition feature extraction network can judge changes in soil moisture in real time by extracting information such as soil moisture content and osmotic impedance, and adjust the irrigation amount in a timely manner. The plant condition feature extraction network can analyze the plant's water requirements by capturing changes in canopy temperature and leaf reflectivity, thereby optimizing water resource use. The environmental condition feature extraction network can monitor the impact of factors such as air temperature and humidity on soil and plants, further improving irrigation decisions. Overall, this real-time, data-driven management approach ensures maximized irrigation efficiency and resource utilization, while reducing water waste and crop growth risks.

[0051] Specifically, such as Figure 2 As shown, the specific steps for analyzing the irrigation demand index of each irrigated area in farmland are as follows: Based on the regional state time series data, analyze the time series characteristic index set of each irrigated area in farmland, including the land, plant, and environmental state time series characteristic indices; based on the time series characteristic index set, analyze the irrigation demand index of each irrigated area in farmland.

[0052] The specific formula for calculating the irrigation demand index of a specific irrigated area within farmland is as follows: ;in, This represents the irrigation demand index for a specific irrigated area within farmland. This refers to the temporal characteristic index of land status in a specific irrigated area within farmland. This refers to the temporal characteristic index of plant status in a certain irrigated area of ​​farmland. These are the plant response coefficients stored in the database. This refers to the temporal characteristic index of the environmental state of a certain irrigated area within farmland. These are the environmental response coefficients stored in the database.

[0053] It needs to be explained that the plant response coefficients stored in the database Environmental response coefficient The specific steps to obtain it are as follows: For plant response coefficients: Select several farmland areas with different soil types and climatic conditions in laboratory or field experiments to ensure that the selected areas can represent diverse soil and climatic conditions. Use sensors to continuously monitor these areas and collect relevant soil data (such as changes in soil moisture, temperature, humidity, etc.). At the same time, for plants, it is necessary to record data such as leaf temperature and near-infrared reflectance at each time point. After preprocessing, the collected data can be used to extract characteristic data that can describe the plant growth status, such as changes in soil moisture content and osmotic resistance. Based on all the collected data, establish a relationship model between plant growth status and water input through regression analysis or other statistical methods, and calculate the plant response coefficient, which represents the plant's sensitivity to changes in soil moisture.

[0054] For the environmental response coefficient: During the experiment, different environmental conditions (such as temperature, humidity, light, etc.) were selected for monitoring. In the experimental area, sensors were used to record soil moisture evaporation, temperature, humidity, light, and other environmental data under different environmental conditions. Through regression analysis of these environmental data and moisture evaporation data, the environmental response coefficient was calculated. This coefficient represents the degree of influence of each environmental factor (such as temperature, humidity, light, etc.) on moisture evaporation. By analyzing the moisture evaporation rate under different environmental conditions, the contribution of different environmental factors to moisture evaporation can be determined, thereby obtaining an accurate environmental response coefficient.

[0055] The specific implementation example for calculating the irrigation demand index of each irrigated area within farmland is as follows, including the time-series characteristic indices of land, plant, and environmental conditions for five irrigated areas. Specific data are shown in Table 1. Table 1. Examples of temporal characteristic index data of land, plants, and environmental conditions in five irrigated areas.

[0056] Furthermore, the plant response coefficient stored in the database is approximately 0.756.

[0057] The environmental response coefficient stored in the database is approximately 1.032.

[0058] Substituting the above data into the specific formula for calculating the irrigation demand index of each irrigated area within the farmland, we obtain the irrigation demand index of the five irrigated areas within the farmland, as follows: Figure 3 As shown: The irrigation demand index of irrigation area 1 in the farmland is 0.652×(1+((1.389)^0.756))×(1+((2.147)^1.032))≈4.761.

[0059] The irrigation demand index of irrigation area 2 in the farmland is 0.736×(1+((1.274)^0.756))×(1+((1.981)^1.032))≈4.900.

[0060] The irrigation demand index of irrigation area 3 in the farmland is 0.812×(1+((1.204)^0.756))×(1+((2.032)^1.032))≈5.376.

[0061] The irrigation demand index of irrigation area 4 in the farmland is 0.675×(1+((1.457)^0.756))×(1+((1.954)^1.032))≈4.711.

[0062] The irrigation demand index of irrigation area 5 in the farmland is 0.799 × (1 + ((1.954)^0.756)) × (1 + ((2.011)^1.032)) ≈ 6.494.

[0063] In this implementation plan, by analyzing the time-series data characteristics of soil, plants, and the environment in detail, the irrigation demand index of each irrigation area can be accurately calculated, avoiding the blind and imprecise irrigation methods of traditional irrigation systems. By introducing plant response coefficients and environmental response coefficients, the system can precisely adjust irrigation strategies according to the impact of different soil and climate conditions on crops. The plant response coefficient helps quantify the sensitivity of crops to changes in soil moisture, while the environmental response coefficient reflects the impact of climate change on water evaporation. Combining these coefficients, the system can more scientifically determine irrigation demand, avoid over-irrigation or under-irrigation, improve water resource utilization efficiency, and enhance crop growth stability and yield. This method makes irrigation decisions more flexible and responsive, ensuring the sustainability of farmland water resource management.

[0064] Specifically, the steps for analyzing the temporal characteristic index of land state in each irrigated area within farmland are as follows: read the slope value of water content change, the standard deviation value of osmotic impedance, and the amplitude value of soil temperature change for each irrigated area within farmland, and perform standardization processing (i.e., unit removal); based on the standardized slope value of water content change, the standard deviation value of osmotic impedance, and the amplitude value of soil temperature change, analyze the temporal characteristic index of land state in each irrigated area within farmland.

[0065] The specific formula for calculating the land state temporal characteristic index of a certain irrigated area within farmland is as follows: ;in, This refers to the temporal characteristic index of land status in a specific irrigated area within farmland. This represents the slope value of the moisture content change in a specific irrigated area within a standardized farmland. The soil-plant interaction coefficients are stored in the database. This represents the standard deviation of osmotic resistance in a specific irrigated area of ​​farmland after standardization. This refers to the influence coefficient of moisture change stored in the database. This represents the temperature variation range of a specific irrigated area within farmland after standardization. This refers to the combined influence coefficient of soil and vegetation stored in the database.

[0066] It needs to be explained that the soil-plant interaction coefficients stored in the database Moisture variation influence coefficient Comprehensive influence coefficient of soil and vegetation The specific steps to obtain it are as follows: For the soil-plant interaction coefficient: collect soil-related characteristic data, such as soil moisture content, temperature, and permeability, and at the same time obtain plant status data, such as leaf reflectance and growth status. These data can be obtained through sensors, remote sensing technology, or field experiments. Calculate the interaction terms between soil and plants, that is, multiply soil status (such as moisture content) with plant status (such as leaf reflectance). Through regression analysis or machine learning models, fit these interaction terms with the plant growth data to obtain the soil-plant interaction coefficient.

[0067] For the influence coefficient of water change: Collect data related to water change, including soil moisture change data (such as water content, infiltration rate, etc.) and plant response data (such as plant growth status, leaf reflectance, etc.). These data are usually obtained through soil sensors, plant monitoring, and remote sensing data. By analyzing these data, the rate of water change is calculated. The calculation of the rate of water change is usually based on the function ln(1+DwB), where DwB represents the impact of water change on plant growth. Through regression analysis or other fitting methods, the influence coefficient of water change is calculated using historical data. This coefficient reflects the impact of water change on plant growth.

[0068] For the soil-plant integrated impact coefficient: collect data related to soil and plant conditions, including soil moisture content, temperature and other characteristics, as well as plant leaf reflectance, growth status and other data. These data can be obtained through various means such as soil sensors, plant growth monitoring and remote sensing equipment. Perform comprehensive analysis on these data, and calculate the comprehensive effect of soil and plant conditions through mathematical operations such as weighted average or square root. Then, fit the comprehensive effect of soil and plant conditions to the plant growth status through regression analysis or other statistical methods to obtain the soil-plant integrated impact coefficient.

[0069] In this implementation plan, by combining factors such as soil, plant, and water changes for comprehensive analysis, more accurate and dynamic irrigation decisions can be made. By calculating the land condition time-series characteristic index, the system can comprehensively assess the combined effects of factors such as soil moisture content changes, permeability, and temperature fluctuations. Using standardized data and combining it with coefficients stored in the database, the system can make flexible irrigation adjustments under different soil conditions. The introduction of soil-plant interaction coefficient, water change coefficient, and soil-plant comprehensive influence coefficient further enhances the system's understanding of the interaction between soil and plants, ensuring the rationality of water allocation. This approach not only improves water resource utilization efficiency and avoids over-irrigation and water waste, but also optimizes the crop growth environment, improving the efficiency and sustainability of agricultural production.

[0070] Specifically, the steps for analyzing the temporal characteristic index of plant status in each irrigated area of ​​farmland are as follows: Read the canopy temperature difference range and leaf reflectance change rate of each irrigated area in the farmland, and perform standardization processing (i.e., unit removal); combine the standard deviation of leaf reflectance of each irrigated area in the farmland with the standardized canopy temperature difference range and leaf reflectance change rate of the corresponding irrigated area for comprehensive analysis to obtain the temporal characteristic index of plant status in each irrigated area of ​​the farmland.

[0071] The specific formula for calculating the temporal characteristic index of plant status in a certain irrigated area of ​​farmland is as follows: ;in, This refers to the temporal characteristic index of plant status in a certain irrigated area of ​​farmland. This represents the canopy temperature range in a standardized irrigated area of ​​farmland. The temperature difference influence coefficient is stored in the database. The rate of change in leaf reflectance in a standardized irrigated area of ​​farmland. The reflectance variation influence coefficients are stored in the database. This represents the standard deviation of leaf reflectance in a specific irrigated area of ​​farmland. The reflectance fluctuation impact coefficient is stored in the database. This refers to the interaction coefficient between the crown and the anti-crown system stored in the database.

[0072] It needs to be explained that the temperature difference influence coefficient is stored in the database. Influence coefficient of reflectivity variation Reflectivity fluctuation influence coefficient Coronavirus-reactive interaction coefficient The specific steps to obtain it are as follows: For the temperature difference influence coefficient: Canopy temperature difference data of farmland is collected. This data is usually obtained through temperature sensors or remote sensing technology, representing the temperature difference range between the soil and plant surface, reflecting the intensity of temperature fluctuations. Then, the canopy temperature difference data is correlated with plant growth data (such as leaf reflectance, growth rate, etc.). Regression analysis or machine learning methods are used to fit the relationship between temperature difference changes and plant growth status, and the temperature difference influence coefficient is calculated. This coefficient represents the influence of canopy temperature difference on the temporal characteristics of plant condition.

[0073] Regarding the reflectance change influence coefficient: Reflectance change rate data from farmland is collected, typically through remote sensing technology or ground sensors. The reflectance change rate reflects changes in light reflected from the plant surface and is usually related to moisture, light conditions, and plant health. Then, this reflectance change rate data is correlated with other plant growth data (such as leaf reflectance, temperature difference, etc.), and regression analysis or other statistical methods are used to fit the relationship between reflectance change and plant growth status. The reflectance change influence coefficient is obtained through this fitting process. This coefficient represents the effect of reflectance changes on the temporal characteristics of plant state.

[0074] For the reflectance fluctuation impact coefficient: Reflectance standard deviation data from farmland is collected. This data is typically acquired using remote sensing technology or ground sensors. The standard deviation reflects the range of reflectance value fluctuations, usually indicating the instability or volatility of plant growth. Then, by calculating the reflectance standard deviation, the reflectance fluctuation value at each time point is obtained. Next, these reflectance standard deviation data are correlated with other plant growth data (such as temperature difference, photosynthetic rate, etc.). Regression analysis is used to calculate the impact of reflectance fluctuations on plant growth status, yielding the reflectance fluctuation impact coefficient. It describes the effect of reflectance fluctuations on the temporal characteristics of plant states.

[0075] For the crown-reflection interaction coefficient: Data on the extreme range of canopy temperature difference and the rate of change of reflectance in farmland were collected. These data reflect the effects of temperature difference on plant physiological state and the rate of change of reflectance on plant adaptability, respectively. By calculating the interaction term between temperature difference and the rate of change of reflectance, the combined effect of their influence on plant state was simulated. The interaction term was then correlated with plant growth data, and regression analysis or other fitting methods were used to determine the crown-reflection interaction coefficient. This indicates the effect of the interaction between temperature difference and the rate of change of reflectance on the plant condition.

[0076] In this implementation plan, by comprehensively analyzing multiple indicators such as canopy temperature difference and leaf reflectance changes, the system can more comprehensively assess the plant's growth status and water requirements. After standardization, the system can eliminate the influence of different data dimensions, ensuring that each indicator accurately reflects its impact on the plant's condition. Using parameters such as the temperature difference influence coefficient and reflectance change influence coefficient stored in the database, the system can quantify the impact of temperature difference and reflectance changes on plant health, further improving the accuracy of irrigation decisions. For example, the canopy-reflectance interaction coefficient helps the system understand the interaction between temperature difference and reflectance, thereby accurately judging the plant's sensitivity to environmental changes. This allows for the optimization of water resource allocation, avoiding over-irrigation or under-irrigation, ensuring that plants are in optimal growth conditions, maximizing resource utilization, and improving agricultural production efficiency and crop quality.

[0077] Specifically, the steps for analyzing the environmental state time-series characteristic index of each irrigated area in farmland are as follows: read the humidity change amplitude value and evaporation offset average value of each irrigated area in farmland and perform standardization processing (i.e., unit removal); combine the temperature change slope value of each irrigated area in farmland with the standardized humidity change amplitude value and evaporation offset average value of the corresponding irrigated area for comprehensive analysis (i.e., weighted analysis) to obtain the environmental state time-series characteristic index of each irrigated area in farmland.

[0078] In this implementation plan, by comprehensively analyzing data such as humidity, evaporation, and temperature changes in the farmland environment, the impact of the environment on crop growth can be accurately assessed. The amplitude of humidity changes and the average evaporation offset reflect the fluctuations in environmental humidity and water evaporation. These factors directly affect soil moisture retention and plant water supply. Combining these data with the slope values ​​of temperature changes for weighted analysis helps to reveal the interactions between different environmental factors and their specific impact on irrigation demand. Through this comprehensive analysis, the system can more accurately determine irrigation demand under different environmental conditions, avoid inaccurate irrigation caused by ignoring environmental changes, improve the scientific and rational nature of irrigation decisions, thereby optimizing water resource use and improving agricultural production efficiency.

[0079] Specifically, the steps for implementing irrigation measures in the irrigated areas of farmland are as follows: Obtain irrigation status data for the irrigated areas of farmland, including the actual area value, the difference between the current soil moisture content and the target soil moisture content, the current soil osmotic resistance value and the soil water absorption rate, and analyze the irrigation water volume and irrigation duration; based on the irrigation water volume and irrigation duration, implement irrigation measures for the irrigated areas.

[0080] Wherein, irrigation water volume = actual area value × (target soil moisture content - current soil moisture content) × (1 / current soil permeability resistance value).

[0081] Irrigation duration = Irrigation water volume / Soil water absorption rate.

[0082] In this implementation plan, by calculating data such as soil moisture content, osmotic resistance, and water absorption rate, the system accurately determines the required amount of water and duration for irrigation, thereby achieving precision irrigation. By calculating the irrigation water volume, the system can avoid over- or under-irrigation based on the difference between the actual soil needs and the target moisture content, ensuring that plants receive appropriate water supply. Combining soil osmotic resistance, the system further considers the soil's permeability characteristics, making the irrigation process more consistent with the soil's actual water absorption capacity and reducing water waste. By calculating the irrigation duration, the system ensures that water can penetrate to the deep soil layers within an appropriate time, preventing water evaporation or loss. This irrigation control method based on precise data calculation can significantly improve water resource utilization efficiency, reduce unnecessary waste, and improve crop growth stability and agricultural production benefits.

[0083] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0084] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. An automated irrigation control system for farmland, characterized in that, include: The time-series status acquisition unit acquires time-series data of the status of several irrigated areas in the farmland, including time-series data of land, plants, and environment. The temporal feature extraction unit, based on a pre-trained temporal state feature extraction model, extracts features from the regional state temporal data of each irrigated area in the farmland, and obtains the regional state temporal data of the corresponding irrigated area, including the temporal feature set of land, plants and environment. The judgment and analysis unit analyzes the irrigation demand index of each irrigation area in the farmland based on the regional status time series data, and makes judgment and analysis with the preset irrigation demand interval. The irrigation execution unit considers irrigation areas where the irrigation demand index is within a preset irrigation demand range as areas requiring irrigation and executes irrigation measures.

2. The automated irrigation control system for farmland according to claim 1, characterized in that, The land condition time series data includes soil moisture content, soil osmotic resistance, and soil temperature values ​​at several time points. The plant condition time series data includes canopy infrared temperature and leaf near-infrared reflectance values ​​at several time points. The environmental condition time series data includes air temperature, air humidity, and micro-area evaporation offset values ​​at several time points.

3. The automated irrigation control system for farmland according to claim 1, characterized in that, The land state temporal feature set includes the slope value of water content change, the standard deviation value of osmotic impedance, and the amplitude value of soil temperature change; the plant state temporal feature set includes the canopy temperature difference range value, the rate of change of leaf reflectance, and the standard deviation of leaf reflectance; the environmental state temporal feature set includes the slope value of air temperature change, the amplitude value of humidity change, and the average value of evaporation offset; and the temporal state feature extraction model includes a land state feature extraction network, a plant state feature extraction network, and an environmental state feature extraction network.

4. The automated irrigation control system for farmland according to claim 3, characterized in that, The specific steps to obtain the regional state time series data of the corresponding irrigation area are as follows: In the land state feature extraction network of the temporal state feature extraction model, the land state temporal data of each irrigated area in the farmland is subjected to multi-time point statistical analysis to obtain the land state temporal feature set of the corresponding irrigated area. In the plant state feature extraction network of the time-series state feature extraction model, the plant state time-series data of each irrigation area in the farmland are subjected to multi-time point fluctuation analysis to obtain the plant state time-series feature set of the corresponding irrigation area. In the environmental state feature extraction network of the time-series state feature extraction model, trend extraction and statistical processing are performed on the environmental state time-series data of each irrigation area in the farmland to obtain the environmental state time-series feature set of the corresponding irrigation area.

5. The automated irrigation control system for farmland according to claim 3, characterized in that, The specific steps for analyzing the irrigation demand index of each irrigated area within farmland are as follows: Based on regional state time series data, we analyze the time series characteristic index set of each irrigated area in farmland, including land, plant and environmental state time series characteristic indices; Based on the time-series characteristic index set, the irrigation demand index of each irrigated area in farmland is analyzed.

6. The automated irrigation control system for farmland according to claim 5, characterized in that, The specific formula for calculating the irrigation demand index of a specific irrigated area within farmland is as follows: ； in, , , , The indices are, in order, the irrigation demand index, the land condition time-series characteristic index, the plant condition time-series characteristic index, and the environmental condition time-series characteristic index for a specific irrigated area within farmland. , The numbers represent, in order, the plant response coefficients and the environmental response coefficients stored in the database.

7. The automated irrigation control system for farmland according to claim 5, characterized in that, The specific steps for analyzing the temporal characteristic index of land status in each irrigated area within farmland are as follows: The slope of water content change, standard deviation of osmotic impedance, and amplitude of soil temperature change were read for each irrigated area in the farmland and then standardized. Based on the standardized values ​​of the slope of moisture content change, the standard deviation of osmotic impedance, and the amplitude of soil temperature change, the temporal characteristic index of land state in each irrigated area in farmland was analyzed.

8. The automated irrigation control system for farmland according to claim 5, characterized in that, The specific steps for analyzing the temporal characteristic index of plant status in each irrigated area within farmland are as follows: The canopy temperature difference range and leaf reflectance change rate of each irrigated area in the farmland were read and standardized. The standard deviation of leaf reflectance in each irrigated area of ​​the farmland was combined with the range of canopy temperature difference and the rate of change of leaf reflectance in the corresponding irrigated areas after standardization to obtain the temporal characteristic index of plant status in each irrigated area of ​​the farmland.

9. The automated irrigation control system for farmland according to claim 5, characterized in that, The specific steps for analyzing the temporal characteristic indices of the environmental status of each irrigated area within farmland are as follows: Read the humidity change amplitude and evaporation offset average of each irrigated area in the farmland, and then standardize them; The slope of temperature change in each irrigated area within the farmland is combined with the standardized amplitude of humidity change and the average evaporation shift of the corresponding irrigated area for comprehensive analysis to obtain the temporal characteristic index of the environmental state of each irrigated area within the farmland.

10. The automated irrigation control system for farmland according to claim 1, characterized in that, The specific steps for implementing irrigation measures in irrigated areas of farmland are as follows: Obtain irrigation status data for the irrigated areas within farmland, including actual area value, the difference between current soil moisture content and target soil moisture content, current soil osmotic resistance value, and soil water absorption rate, and analyze irrigation water volume and irrigation duration; Irrigation measures are implemented in areas requiring irrigation based on irrigation water volume and irrigation duration.

Images