Water and fertilizer control system and method based on artificial intelligence

Abstract

The invention provides a water and fertilizer control system and method based on artificial intelligence, and the method comprises the steps: extracting a chlorophyll absorption index of a crop in a current growth stage and a vegetation index of a growth environment from a canopy multispectral image of the crop, and carrying out the nutrition diagnosis of the growth environment of the crop based on the chlorophyll absorption index and the vegetation index. Obtaining the nutrition stress state of each nutrient element, and determining the target fertilizer solution concentration of each nutrient element according to each nutrition stress state; determining the crop coefficient of the crop in the current growth stage, and performing moisture balance on the crop growth environment according to the crop coefficient, the moisture distribution characteristic in the soil moisture content data and the rainfall evaporation characteristic in the meteorological information of the crop growth environment to obtain the net water demand of the crop in the current growth stage; and generating a fertilization control instruction for directly driving the water and fertilizer integrated equipment based on each target fertilizer solution concentration and the net water demand. Based on the scheme, collaborative decision-making of water and fertilizer supply and demand in a crop growth environment can be realized.

Description

Technical Field

[0001] This application relates to the field of fertilization control technology, and more specifically, to an artificial intelligence-based water and fertilizer control system and method. Background Technology

[0002] Proper irrigation ensures crops receive the water they need for growth, preventing drought or waterlogging. Scientific fertilization provides sufficient nutrients, promoting vigorous plant growth and increasing yield and quality. Precise fertigation technology enables simultaneous water and fertilizer supply, improving resource utilization efficiency and representing an important direction for modern agricultural development.

[0003] Traditional water and fertilizer management typically uses single-point field measurements to set uniform fertilizer formulas and dosages for the entire planting area. This fails to identify and respond to differences in nutrient requirements and spatial stress, meaning that the required proportions of nutrients such as nitrogen, phosphorus, and potassium change dynamically at different growth stages, and fixed formulas are ill-suited to this dynamic nutrient balance. Due to variations in soil texture, microbial activity, and the unevenness of previous fertilization, nutrient availability and crop stress levels differ significantly across field locations. This leads to excessive nutrient accumulation in low-stress areas, causing environmental pollution and resource waste, while persistent nutrient deficiency in high-stress areas restricts yield formation, ultimately resulting in overall water and fertilizer use efficiency significantly lower than theoretically optimal values. Therefore, achieving coordinated decision-making regarding water and fertilizer supply and demand in the crop growth environment has become a major challenge for the industry. Summary of the Invention

[0004] This application provides an artificial intelligence-based water and fertilizer control system and method, which can realize collaborative decision-making on water and fertilizer supply and demand in the crop growth environment.

[0005] In a first aspect, this application provides a water and fertilizer control method based on artificial intelligence, comprising: Acquire soil moisture data and meteorological information of the crop growth environment, and acquire canopy multispectral images of crops using multispectral sensors; The chlorophyll absorption index of crops at the current growth stage and the vegetation index of the growth environment are extracted from the canopy multispectral image. Based on the chlorophyll absorption index and the vegetation index, the nutrient environment of the crop growth is diagnosed to obtain the nutrient stress status of each nutrient element. The target fertilizer concentration of each nutrient element is determined by the nutrient stress status and the nutrient distribution characteristics of the crop growth environment. Determine the crop coefficient at the current growth stage of the crop, and perform water balance analysis on the crop growth environment based on the crop coefficient, the water distribution characteristics in the soil moisture data, and the rainfall and evaporation characteristics in the meteorological information to obtain the net water requirement of the crop at the current growth stage. Based on the target fertilizer concentration and the net water requirement, fertilization control commands are generated to directly drive the integrated water and fertilizer equipment.

[0006] In some embodiments, extracting the chlorophyll uptake index of crops at their current growth stage and the vegetation index of the growing environment from the canopy multispectral image specifically includes: The canopy multispectral image is preprocessed to obtain the surface reflectance image of the crop canopy; Based on the surface reflectance image, the chlorophyll absorption index of crops at the current growth stage and the vegetation index of the growth environment are calculated.

[0007] In some embodiments, nutritional diagnosis of the crop growth environment based on the chlorophyll absorption index and the vegetation index, to obtain the nutritional stress status of each nutrient element, specifically includes: An input layer is constructed with the chlorophyll absorption index, the vegetation index, and the current growth stage of the crop as feature vectors; The feature vector is input into a pre-trained multilayer perceptron classification model using the input layer. The multilayer perceptron classification model is used to classify and diagnose the crop growth environment, and to obtain the nutritional stress status of each nutrient element.

[0008] In some embodiments, determining the target fertilizer concentration for each nutrient element based on the nutrient distribution characteristics of various nutrient stress states and crop growth environments specifically includes: For each nutrient element, the effective content of the nutrient element in the soil is obtained from the nutrient distribution characteristics of the crop growth environment, and then the nutrient correction coefficient of the soil for the nutrient element is determined. Based on the nutrient stress status of nutrient elements, the basic fertilization concentration and stress correction coefficient of nutrient elements can be retrieved from a predefined decision table; The target fertilizer solution concentrations of each nutrient element are determined based on the nutrient correction coefficient, the basic fertilizer concentration, and the stress correction coefficient, thereby obtaining the target fertilizer solution concentrations of each nutrient element.

[0009] In some embodiments, determining the crop coefficient at the current growth stage of the crop specifically includes: Obtain the vegetation index of the crop growth environment; Based on the crop type of the crop, obtain the basic crop coefficient of the crop at the current growth stage; The basic crop coefficient is corrected by the vegetation index to obtain the crop coefficient of the crop at the current growth stage.

[0010] In some embodiments, the water balance of the crop growth environment is calculated based on the crop coefficient, the water distribution characteristics in the soil moisture data, and the rainfall-evaporation characteristics in the meteorological information to obtain the net water requirement of the crop at the current growth stage, specifically including: The actual water requirement of crops is determined by the crop coefficient and the evapotranspiration of the crop growing environment. Effective rainfall data for crop growth environment can be obtained from rainfall and evaporation characteristics in meteorological information. The surplus or deficit of soil moisture in the crop growth environment is obtained from the water distribution characteristics in soil moisture data. The net water requirement of crops at the current growth stage is determined by the actual water demand, the effective rainfall, and the profit / loss.

[0011] In some embodiments, generating fertilization control instructions for directly driving the fertigation equipment based on each target fertilizer solution concentration and the net water requirement specifically includes: Based on the net water demand and the maximum single irrigation efficiency of the integrated water and fertilizer equipment, calculate the planned opening time of the main pipeline irrigation valve in the integrated water and fertilizer equipment. Calculate the injection flow rate of each mother liquor pump in the integrated water and fertilizer system based on the target fertilizer concentration and the rated flow rate of the integrated water and fertilizer system. The planned start-up duration and the injection flow rate of each mother liquor pump are combined to generate fertilization control commands that directly drive the integrated water and fertilizer equipment.

[0012] Secondly, this application provides an artificial intelligence-based water and fertilizer control system, comprising: The acquisition module is used to acquire soil moisture data and meteorological information of the crop growth environment, and to acquire canopy multispectral images of crops collected using a multispectral sensor; The processing module is used to extract the chlorophyll absorption index of crops at the current growth stage and the vegetation index of the growth environment from the canopy multispectral image, perform nutrient diagnosis on the crop growth environment based on the chlorophyll absorption index and the vegetation index, obtain the nutrient stress status of each nutrient element, and determine the target fertilizer concentration of each nutrient element through the nutrient stress status and the nutrient distribution characteristics of the crop growth environment. The processing module is also used to determine the crop coefficient of the crop at the current growth stage, and to perform water balance on the crop growth environment based on the crop coefficient, the water distribution characteristics in the soil moisture data and the rainfall and evaporation characteristics in the meteorological information, so as to obtain the net water requirement of the crop at the current growth stage. The execution module is used to generate fertilization control commands that directly drive the integrated water and fertilizer equipment based on each target fertilizer solution concentration and the net water requirement.

[0013] Thirdly, this application provides a computer device, which includes a memory and a processor. The memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that the computer device executes the above-described artificial intelligence-based water and fertilizer control method.

[0014] Fourthly, this application provides a computer-readable storage medium storing instructions or code that, when executed on a computer, cause the computer to implement the aforementioned artificial intelligence-based water and fertilizer control method.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: This application provides an artificial intelligence-based water and fertilizer control system and method, which acquires soil moisture data and meteorological information of the crop growth environment, and acquires canopy multispectral images of the crop using a multispectral sensor; extracts the chlorophyll absorption index of the crop at the current growth stage and the vegetation index of the growth environment from the canopy multispectral images; performs nutrient diagnosis on the crop growth environment based on the chlorophyll absorption index and the vegetation index to obtain the nutrient stress state of each nutrient element; determines the target fertilizer concentration of each nutrient element based on the nutrient stress state and the nutrient distribution characteristics of the crop growth environment; determines the crop coefficient of the crop at the current growth stage; performs water balance on the crop growth environment based on the crop coefficient, the water distribution characteristics in the soil moisture data, and the rainfall-evaporation characteristics in the meteorological information to obtain the net water requirement of the crop at the current growth stage; and generates fertilization control instructions to directly drive the integrated water and fertilizer equipment based on the target fertilizer concentrations and the net water requirement.

[0016] Therefore, in this application, fertilization control instructions for the integrated water and fertilizer system are generated based on the target fertilizer concentration and the net water requirement. First, determining the target fertilizer concentration yields a fertilization prescription that matches the crop's nutrient needs with the soil's baseline supply, thus realizing a shift in fertilization decisions from homogeneous constants to on-demand variables. It can accurately identify whether the crop is under single or combined nitrogen, phosphorus, or potassium stress, and performs secondary corrections based on the soil's baseline nutrients, outputting a unique target concentration for each nutrient element. This ensures that the nutrient composition in the irrigation solution directly responds to the crop's immediate nutrient deficiency signal and effectively avoids ineffective fertilization in eutrophic soil areas. It solves the core defects of traditional fertilization, such as fixed nutrient ratios and inability to adapt to spatiotemporal variability, providing direct data-driven instructions for precise variable fertilization. Then, determining... The net water requirement allows for irrigation that meets the physiological water needs of crops while also taking into account environmental water supply, thus transforming irrigation decisions. This enables irrigation practices to proactively adapt to weather changes and soil moisture retention capacity, avoiding water leakage, runoff loss, and nutrient leaching caused by rigid irrigation systems. This not only significantly improves water use efficiency but also ensures that applied fertilizer is retained to the maximum extent possible in the crop root zone, creating the necessary water environment for efficient fertilizer utilization. Finally, the target fertilizer concentration ensures that the nutrients dissolved in the water are those urgently needed by the crops, while the net water requirement ensures that the water carrying the nutrients is used efficiently. Together, they form a closed-loop intelligent control system centered on crop needs and aiming for resource efficiency, ultimately achieving precise and coordinated decision-making regarding water and fertilizer supply and demand in time, space, and quantity. In summary, the above solution enables coordinated decision-making regarding water and fertilizer supply and demand in the crop growth environment. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is an exemplary flowchart of an artificial intelligence-based water and fertilizer control method according to some embodiments of this application; Figure 2 This is a flowchart illustrating the process of determining crop coefficients according to some embodiments of this application; Figure 3 This is a schematic diagram of the structure of an artificial intelligence-based water and fertilizer control system according to some embodiments of this application; Figure 4This is a schematic diagram of the structure of a computer device that implements an artificial intelligence-based water and fertilizer control method according to some embodiments of this application. Detailed Implementation

[0019] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] refer to Figure 1 The figure is an exemplary flowchart of an artificial intelligence-based water and fertilizer control method according to some embodiments of this application. The artificial intelligence-based water and fertilizer control method mainly includes the following steps: In step 101, soil moisture data and meteorological information of the crop growth environment are acquired, and canopy multispectral images of the crops are acquired using a multispectral sensor.

[0021] It should be noted that, in this application, soil moisture data is a set of physical parameters characterizing the content and distribution of soil moisture in the root zone of crops; meteorological information is a set of environmental parameters reflecting the atmospheric physical state and changing trends of the crop growth environment; and canopy multispectral images are digital images used to analyze the physiological state and growth of crops.

[0022] In specific implementation, firstly, obtaining soil moisture data and meteorological information of the crop growth environment can be achieved in the following ways: Wireless soil sensors buried in the main root layer of crops periodically measure and upload soil volumetric water content data and soil temperature data, using the collection of soil volumetric water content data and soil temperature data as soil moisture data of the crop growth environment; Meteorological sensors installed in field weather stations collect environmental data including air temperature, air humidity, wind speed, sunshine duration, and rainfall, thus using this environmental data as meteorological information of the crop growth environment; Then, obtaining multispectral images of the crop canopy collected using multispectral sensors can be achieved in the following ways: Multispectral sensors vertically downwards calibrate and photograph the crop canopy to obtain raw digital images containing visible light and near-infrared reflectance information, using this raw digital image as the canopy multispectral image. This canopy multispectral image can be stored in a data center, and the crop canopy multispectral image collected by the multispectral sensor can be obtained from the data center via, for example, wireless communication. No specific limitations are made on the acquisition method here.

[0023] In step 102, the chlorophyll absorption index of the crop at the current growth stage and the vegetation index of the growth environment are extracted from the canopy multispectral image. Based on the chlorophyll absorption index and the vegetation index, the crop growth environment is nutrient-dependently diagnosed to obtain the nutrient stress status of each nutrient element. The target fertilizer concentration of each nutrient element is determined by the nutrient stress status and the nutrient distribution characteristics of the crop growth environment.

[0024] In some embodiments, extracting the chlorophyll uptake index of crops at their current growth stage and the vegetation index of the growing environment from the canopy multispectral image can be achieved by the following steps: The canopy multispectral image is preprocessed to obtain the surface reflectance image of the crop canopy; Based on the surface reflectance image, the chlorophyll absorption index of crops at the current growth stage and the vegetation index of the growth environment are calculated.

[0025] It should be noted that in this application, the chlorophyll absorption index is a remote sensing spectral index that quantitatively assesses the relative concentration of chlorophyll inside crop leaves and the nitrogen nutrition status; the vegetation index is a remote sensing spectral index that reflects the canopy coverage, biomass and growth vitality of crops; and the surface reflectance image is a corrected image that characterizes the true reflectivity of ground features to solar radiation.

[0026] In specific implementation, firstly, the canopy multispectral image is preprocessed to obtain the surface reflectance image of the crop canopy. This can be achieved as follows: using the calibration coefficient file included with the machine vision sensor in the fertigation equipment, the canopy multispectral image is radiometrically calibrated, converting the image pixel values ​​into physically meaningful atmospheric top-layer reflectance; an atmospheric correction model is invoked, combined with meteorological data acquired synchronously during image acquisition, to eliminate the scattering and absorption effects of atmospheric components such as aerosols and water vapor from the atmospheric top-layer reflectance, thereby geometrically cropping the atmospherically corrected image to retain only the effective canopy area of ​​the target crop. The image, after the above preprocessing, representing the true spectral properties of the crop canopy, is then used as the surface reflectance image of the crop canopy; finally, based on the surface reflectance image, the crop's current growth stage is calculated. The chlorophyll absorption index and the vegetation index of the growing environment can be achieved as follows: Read the reflectance values ​​of the crop canopy in the red light band, near-infrared band, and a designated red-edge band sensitive to chlorophyll from a surface reflectance image; Substitute the red-edge band reflectance and red light band reflectance into the calculation based on the definition formula of the chlorophyll absorption index, i.e., chlorophyll absorption index = (red-edge band reflectance - red light band reflectance) / (red-edge band reflectance + red light band reflectance), and use the result as the chlorophyll absorption index of the crop at the current growth stage; Simultaneously, according to the definition formula of the vegetation index, i.e., vegetation index = (near-infrared band reflectance - red light band reflectance) / (near-infrared band reflectance + red light band reflectance), substitute the near-infrared band reflectance and red light band reflectance into the calculation, and use the result as the vegetation index of the crop at the current growth stage.

[0027] In some embodiments, the nutritional stress status of each nutrient element can be obtained by performing nutritional diagnosis of the crop growth environment based on the chlorophyll absorption index and the vegetation index using the following steps: An input layer is constructed with the chlorophyll absorption index, the vegetation index, and the current growth stage of the crop as feature vectors; The feature vector is input into a pre-trained multilayer perceptron classification model using the input layer. The multilayer perceptron classification model is used to classify and diagnose the crop growth environment, and to obtain the nutritional stress status of each nutrient element.

[0028] It should be noted that in this application, the nutrient stress state is a discrete classification result used to quantitatively describe the various nutrient elements necessary for crops; the feature vector is a unified and standardized representation of the combination of observation indicators characterizing different aspects of crop status; the multilayer perceptron classification model is a discriminative classifier built based on artificial neural networks. This multilayer perceptron classification model achieves high-level abstraction and pattern recognition of input features through multilayer nonlinear transformations. The model consists of an input layer, at least one hidden layer, and an output layer. Each layer contains several neurons, and the layers are fully connected through adjustable weight matrices and bias vectors. During forward propagation, the feature vector of the input layer is first linearly combined with the weight matrix of the first hidden layer, and then nonlinearly mapped by an activation function (e.g., the sigmoid activation function). This process is iterated layer by layer in subsequent hidden layers to gradually extract deep features for classification. The output layer uses the Softmax function to convert the output of the hidden layer into a probability distribution vector, where each node corresponds to the probability of each category (e.g., deficiency, normal, excess) of a nutrient element (e.g., nitrogen, phosphorus, potassium). The parameters of the entire model (weights and biases) are pre-optimized through supervised learning on historical data (e.g., backpropagation algorithm), thereby establishing a complex nonlinear mapping relationship from multi-source features to nutrient stress state, thus achieving end-to-end intelligent diagnosis.

[0029] In specific implementation, firstly, the input layer, which uses the chlorophyll absorption index, the vegetation index, and the crop's current growth stage code as feature vectors, can be constructed in the following way: Based on a preset crop phenological period division standard, the crop's current growth stage (e.g., seedling stage, jointing stage, heading stage) is converted into a single effective code; the chlorophyll absorption index, vegetation index, and effective code are concatenated and combined in a predefined fixed order, thus using the ordered array formed by these three indices as the standardized feature vector input to the classification model; then, the input layer is used to input the feature vector into the pre-trained multilayer perceptron classification model, which can be implemented in the following way: the input layer acts as the data interface for the multilayer perceptron classification model, receiving the feature vector; this feature vector is... The input data is directly passed to the first layer neurons of the pre-trained multilayer perceptron classification model. The pre-set weight matrix and bias vector within the multilayer perceptron classification model then begin to work, passing the input data to the output layer according to the forward propagation rules, thus using the input feature vector as the starting point for the model's forward propagation calculation. Finally, based on the multilayer perceptron classification model, the crop growth environment is classified and diagnosed. The nutrient stress state of each nutrient element can be obtained in the following way: the received feature vector is subjected to layer-by-layer nonlinear transformation, thereby generating a set of probability distributions corresponding to the quantity of the target nutrient element in the output layer; by selecting the category with the highest probability value in each output node as the nutrient stress state of the corresponding nutrient element, the nutrient stress state of each nutrient element can be obtained.

[0030] In some embodiments, determining the target fertilizer concentration for each nutrient element based on the nutrient distribution characteristics of various nutrient stress states and crop growth environments can be achieved through the following steps: For each nutrient element, the effective content of the nutrient element in the soil is obtained from the nutrient distribution characteristics of the crop growth environment, and then the nutrient correction coefficient of the soil for the nutrient element is determined. Based on the nutrient stress status of nutrient elements, the basic fertilization concentration and stress correction coefficient of nutrient elements can be retrieved from a predefined decision table; The target fertilizer solution concentrations of each nutrient element are determined based on the nutrient correction coefficient, the basic fertilizer concentration, and the stress correction coefficient, thereby obtaining the target fertilizer solution concentrations of each nutrient element.

[0031] It should be noted that, in this application, the target fertilizer concentration is the final fertilizer concentration value used to guide the fertilizer ratio of the fertigation equipment; the nutrient correction coefficient is a proportional factor characterizing the reduction or supplementation effect of the effective nutrients contained in the soil itself on the fertilizer concentration that needs to be applied additionally; the basic fertilizer concentration is a reference value of fertilizer concentration that can meet basic growth needs; and the stress correction coefficient is a proportional adjustment factor that amplifies or reduces the basic fertilizer concentration.

[0032] In practice, firstly, for each nutrient element, the effective content of the nutrient element in the soil is obtained from the nutrient distribution characteristics of the crop growth environment. The nutrient correction coefficient of the soil for each nutrient element can then be determined in the following way: Sampling points are set up in the farmland according to standardized soil sampling methods. Soil samples are collected at each target tillage layer depth using a soil sampling drill. The samples are sent to a laboratory and analyzed according to national standard methods to obtain the nutrient distribution characteristics of the crop growth environment. From these characteristics, the measured values ​​of the effective content of the target nutrient element (e.g., available nitrogen, available phosphorus, and available potassium) within the specified tillage layer depth are read. These measured values ​​are then substituted into a preset empirical function that reflects the negative correlation between soil nutrient supply capacity and fertilizer concentration for calculation. The calculated value, ranging from 0 to 1, is then used as the nutrient correction coefficient for that nutrient element. The corresponding nutrient correction coefficient is determined. Then, based on the nutrient stress state of the nutrient elements, the basic fertilization concentration and stress correction coefficient of the nutrient elements can be retrieved from the predefined decision table in the following way: a predefined decision table is constructed with different nutrient stress states (e.g., "deficiency", "normal", "excess") as rows and corresponding fertilization parameters as columns. The basic fertilization concentration and stress correction coefficient of the nutrient elements are located and read in the target row of the decision table according to the nutrient stress state of the nutrient elements. Finally, the target fertilizer solution concentration of the nutrient elements is determined based on the nutrient correction coefficient, the basic fertilization concentration, and the stress correction coefficient. The target fertilizer solution concentration of each nutrient element can be obtained in the following way: the product of the nutrient correction coefficient, the basic fertilization concentration, and the stress correction coefficient is taken as the target fertilizer solution concentration of the nutrient element. The target fertilizer solution concentration of each nutrient element can be obtained in the above way.

[0033] In step 103, the crop coefficient of the crop at the current growth stage is determined. Based on the crop coefficient, the water distribution characteristics in the soil moisture data and the rainfall and evaporation characteristics in the meteorological information, the water balance of the crop growth environment is performed to obtain the net water requirement of the crop at the current growth stage.

[0034] In some embodiments, the crop coefficient for the crop at its current growth stage is determined, with reference to... Figure 2 The diagram is a flowchart illustrating the determination of crop coefficients in some embodiments of this application. In this embodiment, the determination of crop coefficients can be achieved using the following steps: In step 1031, the vegetation index of the crop growth environment is obtained; In step 1032, based on the crop type of the crop, the basic crop coefficient of the crop at the current growth stage is obtained; In step 1033, the basic crop coefficient is corrected by the vegetation index to obtain the crop coefficient of the crop at the current growth stage.

[0035] It should be noted that in this application, the crop coefficient reflects the correction coefficient for the actual water requirement of the crop at present; the vegetation index is a remote sensing spectral index that characterizes the canopy coverage and growth vitality of crops; and the basic crop coefficient is an empirical reference value that characterizes the proportion of water requirement of crops at the current growth stage relative to the evapotranspiration of reference crops under standard conditions.

[0036] In specific implementation, firstly, the vegetation index of the crop's growth environment is obtained; then, based on the crop type, the basic crop coefficient of the crop at the current growth stage can be obtained in the following way: a standard crop coefficient table published by an authoritative agricultural institution corresponding to the crop type is consulted. This standard crop coefficient table defines the reference values ​​of the coefficient for the crop at different typical growth stages (e.g., early, middle, and late stages). The corresponding value is located from the standard crop coefficient table based on the current growth stage of the crop as the basic crop coefficient for that stage. Finally, the basic crop coefficient is corrected using the vegetation index to obtain the crop coefficient at the current growth stage. This can be achieved in the following way: the vegetation index is used as a real-time correction factor reflecting the actual canopy coverage of the crop, and it is substituted together with the basic crop coefficient into a preset linear correction formula for calculation. This linear correction formula expresses a positive correlation between the crop coefficient and the increase of the vegetation index. The result calculated by the correction formula is used as the crop coefficient at the current growth stage.

[0037] In some embodiments, the net water requirement of crops at the current growth stage can be obtained by performing water balance analysis on the crop growth environment based on the crop coefficient, water distribution characteristics in soil moisture data, and rainfall-evaporation characteristics in meteorological information, using the following steps: The actual water requirement of crops is determined by the crop coefficient and the evapotranspiration of the crop growing environment. Effective rainfall data for crop growth environment can be obtained from rainfall and evaporation characteristics in meteorological information. The surplus or deficit of soil moisture in the crop growth environment is obtained from the water distribution characteristics in soil moisture data. The net water requirement of crops at the current growth stage is determined by the actual water demand, the effective rainfall, and the profit / loss.

[0038] It should be noted that in this application, net water requirement is the amount of pure water that needs to be supplemented to crops through irrigation; actual water requirement is an estimated value of the total amount of water consumed to maintain normal growth and development; effective rainfall is an estimated value of rainfall that can be actually absorbed and utilized by the crop root system soil layer; and surplus / deficit is the difference between the actual soil moisture content in the planned wetting layer of the crop and the optimal moisture content for crop growth.

[0039] In specific implementation, firstly, determining the actual water requirement of crops based on the crop coefficient and the evapotranspiration of the crop growing environment can be achieved by: obtaining the reference crop evapotranspiration calculated using the standard Penman formula; multiplying the crop coefficient by the reference crop evapotranspiration as the actual water requirement of the crop at the current growth stage; secondly, obtaining the effective rainfall of the crop growing environment from the rainfall-evaporation characteristics in meteorological information can be achieved by: obtaining the total rainfall during the calculation period from meteorological information; substituting this total rainfall into a preset empirical coefficient method model (e.g., the method recommended by the USDA Soil Conservation Service) for calculation; this empirical coefficient method model uses coefficients to characterize the portion of the total rainfall effectively retained by the soil; finally, the result calculated by the empirical coefficient method model is taken as the effective rainfall that can be utilized by crops; then, obtaining the soil moisture surplus / deficit of the crop growing environment from the water distribution characteristics in soil moisture data. The deficit can be calculated as follows: Obtain the average volumetric water content within the main root activity layer (i.e., the planned wetting layer) from soil moisture data. Compare this average volumetric water content with the optimal field capacity for crop growth under this soil type and calculate the difference. Use this difference as the soil moisture surplus / deficit in the crop's growth environment. Finally, determine the net water requirement of the crop at the current growth stage using the actual water demand, effective rainfall, and the surplus / deficit. This can be achieved as follows: Based on the principle of farmland water balance, take the actual water demand as the total consumption, subtract the effective rainfall to deduct natural replenishment, and then subtract the soil moisture surplus / deficit to consider the soil's own regulation capacity. A positive surplus / deficit indicates that the soil has water available for crop use, so it is subtracted; a negative surplus / deficit indicates that the soil itself is short of water and needs additional replenishment. The result of the water balance calculation is then used as the net water requirement of the crop at the current growth stage.

[0040] In step 104, fertilization control instructions for directly driving the fertigation equipment are generated based on each target fertilizer concentration and the net water requirement.

[0041] In some embodiments, generating fertilization control commands for directly driving the fertigation equipment based on each target fertilizer solution concentration and the net water requirement can be achieved through the following steps: Based on the net water demand and the maximum single irrigation efficiency of the integrated water and fertilizer equipment, calculate the planned opening time of the main pipeline irrigation valve in the integrated water and fertilizer equipment. Calculate the injection flow rate of each mother liquor pump in the integrated water and fertilizer system based on the target fertilizer concentration and the rated flow rate of the integrated water and fertilizer system. The planned start-up duration and the injection flow rate of each mother liquor pump are combined to generate fertilization control commands that directly drive the integrated water and fertilizer equipment.

[0042] In specific implementation, firstly, based on the net water demand and the maximum single-cycle irrigation efficiency of the integrated water and fertilizer system, the planned opening time of the main pipeline irrigation valve in the integrated water and fertilizer system can be calculated as follows: obtain the rated flow rate specified in the technical parameters of the integrated water and fertilizer system and the maximum single-cycle irrigation efficiency determined through testing or experience; divide the net water demand by (the product of the rated flow rate and the maximum single-cycle irrigation efficiency); and use the quotient obtained from the division as the planned opening time of the main pipeline irrigation valve in the integrated water and fertilizer system. Then, based on the target fertilizer concentration and the rated flow rate of the integrated water and fertilizer system, the injection flow rate of each mother liquor pump in the integrated water and fertilizer system can be calculated as follows: obtain the rated flow rate specified in the technical parameters of the integrated water and fertilizer system and the concentration of each nutrient element's mother liquor stock solution; for each nutrient element, multiply the target fertilizer concentration of the nutrient element by the rated flow rate. Then, divide the resulting product by the concentration of the nutrient element's mother liquor stock solution, and use the quotient obtained from the division as the injection flow rate of the corresponding mother liquor pump for the nutrient element. This method yields the injection flow rate of each nutrient element's corresponding mother liquor pump, thus providing the injection flow rate of each mother liquor pump in the integrated water and fertilizer system. Finally, combining the planned start-up duration and the injection flow rate of each mother liquor pump to generate a fertilization control command that directly drives the integrated water and fertilizer system can be achieved in the following way: convert the planned start-up duration into "open" and "close" commands for the irrigation valve controller, and set the corresponding time parameters; simultaneously, convert the injection flow rate of each mother liquor pump into a frequency or speed setting command for the corresponding mother liquor pump's frequency converter or stepper motor; and use the set of operation commands containing timing and flow setting parameters as a fertilization control command sent to the local controller of the integrated water and fertilizer system, which can directly drive the equipment to execute.

[0043] Furthermore, in another aspect of this application, in some embodiments, this application provides a water and fertilizer control system based on artificial intelligence, referencing... Figure 3 The figure is a schematic diagram of the structure of an artificial intelligence-based water and fertilizer control system according to some embodiments of this application. The artificial intelligence-based water and fertilizer control system includes: an acquisition module 201, a processing module 202, and an execution module 203, which are described below: The acquisition module 201 in this application is mainly used to acquire soil moisture data and meteorological information of the crop growth environment, and to acquire canopy multispectral images of crops collected using a multispectral sensor. Processing module 202 in this application is used to extract the chlorophyll absorption index of crops at the current growth stage and the vegetation index of the growth environment from the canopy multispectral image, perform nutrient diagnosis on the crop growth environment based on the chlorophyll absorption index and the vegetation index, obtain the nutrient stress state of each nutrient element, and determine the target fertilizer concentration of each nutrient element through the nutrient stress state and the nutrient distribution characteristics of the crop growth environment. It should be noted that the processing module 202 is also used to determine the crop coefficient of the crop at the current growth stage, and to perform water balance on the crop growth environment based on the crop coefficient, the water distribution characteristics in the soil moisture data and the rainfall and evaporation characteristics in the meteorological information, so as to obtain the net water requirement of the crop at the current growth stage. The execution module 203 in this application is mainly used to generate fertilization control instructions for directly driving the integrated water and fertilizer equipment based on the concentration of each target fertilizer solution and the net water requirement.

[0044] The foregoing has detailed examples of the artificial intelligence-based water and fertilizer control system and method provided in the embodiments of this application. It is understood that the corresponding device, in order to achieve the above functions, includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0045] In some embodiments, this application also provides a computer device, the computer device including a memory and a processor, the memory for storing a computer program, and the processor for calling and running the computer program from the memory, so that the computer device performs the above-described artificial intelligence-based water and fertilizer control method.

[0046] In some embodiments, reference Figure 4 The dashed lines in the figure indicate that the unit or module is optional. This figure is a schematic diagram of the structure of a computer device implementing an artificial intelligence-based water and fertilizer control method according to an embodiment of this application. The artificial intelligence-based water and fertilizer control method described in the above embodiments can... Figure 4 The computer device shown is used to implement this, and the computer device includes at least one processor 301, a memory 302 and at least one communication unit 305. The computer device may be a terminal device, a server or a chip.

[0047] Processor 301 can be a general-purpose processor or a special-purpose processor. For example, processor 301 can be a central processing unit (CPU), which can be used to control computer devices, execute software programs, and process data from software programs. The computer device may also include a communication unit 305 for inputting (receiving) and outputting (transmitting) signals.

[0048] For example, the computer device may be a chip, and the communication unit 305 may be the input and / or output circuit of the chip, or the communication unit 305 may be the communication interface of the chip, which may be a component of a terminal device, network device or other device.

[0049] For example, the computer device may be a terminal device or a server, and the communication unit 305 may be a transceiver of the terminal device or the server, or the communication unit 305 may be a transceiver circuit of the terminal device or the server.

[0050] The computer device may include one or more memories 302 storing a program 304. The program 304 can be executed by a processor 301 to generate instructions 303, causing the processor 301 to execute the method described in the above method embodiments according to the instructions 303. Optionally, the memory 302 may also store data (such as a target audit model). Optionally, the processor 301 may also read data stored in the memory 302, which may be stored at the same storage address as the program 304, or it may be stored at a different storage address than the program 304.

[0051] The processor 301 and memory 302 can be configured separately or integrated together, for example, integrated on the system on chip (SOC) of the terminal device.

[0052] It should be understood that each step of the above method embodiment can be completed by hardware logic circuits or software instructions in the processor 301. The processor 301 can be a CPU, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, such as discrete gates, transistor logic devices, or discrete hardware components.

[0053] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0054] For example, in some embodiments, this application also provides a computer-readable storage medium storing instructions or code that, when executed on a computer, cause the computer to implement the above-described artificial intelligence-based water and fertilizer control method.

[0055] Although preferred embodiments of this application 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 the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0056] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A water and fertilizer control method based on artificial intelligence, characterized in that, Includes the following steps: Acquire soil moisture data and meteorological information of the crop growth environment, and acquire canopy multispectral images of crops using multispectral sensors; The chlorophyll absorption index of crops at the current growth stage and the vegetation index of the growth environment are extracted from the canopy multispectral image. Based on the chlorophyll absorption index and the vegetation index, the nutrient environment of the crop growth is diagnosed to obtain the nutrient stress status of each nutrient element. The target fertilizer concentration of each nutrient element is determined by the nutrient stress status and the nutrient distribution characteristics of the crop growth environment. Determine the crop coefficient at the current growth stage of the crop, and perform water balance analysis on the crop growth environment based on the crop coefficient, the water distribution characteristics in the soil moisture data, and the rainfall and evaporation characteristics in the meteorological information to obtain the net water requirement of the crop at the current growth stage. Based on the target fertilizer concentration and the net water requirement, fertilization control commands are generated to directly drive the integrated water and fertilizer equipment.

2. The method as described in claim 1, characterized in that, Extracting the chlorophyll uptake index of crops at their current growth stage and the vegetation index of the growing environment from the canopy multispectral images specifically includes: The canopy multispectral image is preprocessed to obtain the surface reflectance image of the crop canopy; Based on the surface reflectance image, the chlorophyll absorption index of crops at the current growth stage and the vegetation index of the growth environment are calculated.

3. The method as described in claim 1, characterized in that, Nutritional diagnosis of the crop growth environment based on the chlorophyll absorption index and the vegetation index yields the specific nutritional stress status of each nutrient element, including: An input layer is constructed with the chlorophyll absorption index, the vegetation index, and the current growth stage of the crop as feature vectors; The feature vector is input into a pre-trained multilayer perceptron classification model using the input layer. The multilayer perceptron classification model is used to classify and diagnose the crop growth environment, and to obtain the nutritional stress status of each nutrient element.

4. The method as described in claim 1, characterized in that, Determining the target fertilizer solution concentration for each nutrient element based on various nutrient stress states and the nutrient distribution characteristics of the crop growth environment specifically includes: For each nutrient element, the effective content of the nutrient element in the soil is obtained from the nutrient distribution characteristics of the crop growth environment, and then the nutrient correction coefficient of the soil for the nutrient element is determined. Based on the nutrient stress status of nutrient elements, the basic fertilization concentration and stress correction coefficient of nutrient elements can be retrieved from a predefined decision table; The target fertilizer solution concentrations of each nutrient element are determined based on the nutrient correction coefficient, the basic fertilizer concentration, and the stress correction coefficient, thereby obtaining the target fertilizer solution concentrations of each nutrient element.

5. The method as described in claim 1, characterized in that, Determining the crop coefficient for a crop at its current growth stage specifically includes: Obtain the vegetation index of the crop growth environment; Based on the crop type of the crop, obtain the basic crop coefficient of the crop at the current growth stage; The basic crop coefficient is corrected by the vegetation index to obtain the crop coefficient of the crop at the current growth stage.

6. The method as described in claim 1, characterized in that, Based on the crop coefficient, the water distribution characteristics in the soil moisture data, and the rainfall-evaporation characteristics in the meteorological information, the water balance of the crop growth environment is calculated to obtain the net water requirement of the crop at the current growth stage, specifically including: The actual water requirement of crops is determined by the crop coefficient and the evapotranspiration of the crop growing environment. Effective rainfall data for crop growth environment can be obtained from rainfall and evaporation characteristics in meteorological information. The surplus or deficit of soil moisture in the crop growth environment is obtained from the water distribution characteristics in soil moisture data. The net water requirement of crops at the current growth stage is determined by the actual water demand, the effective rainfall, and the profit / loss.

7. The method as described in claim 1, characterized in that, The fertilization control instructions generated based on the target fertilizer concentration and the net water requirement for the direct-drive fertigation equipment specifically include: Based on the net water demand and the maximum single irrigation efficiency of the integrated water and fertilizer equipment, calculate the planned opening time of the main pipeline irrigation valve in the integrated water and fertilizer equipment. Calculate the injection flow rate of each mother liquor pump in the integrated water and fertilizer system based on the target fertilizer concentration and the rated flow rate of the integrated water and fertilizer system. The planned start-up duration and the injection flow rate of each mother liquor pump are combined to generate fertilization control commands that directly drive the integrated water and fertilizer equipment.

8. A water and fertilizer control system based on artificial intelligence, characterized in that, include: The acquisition module is used to acquire soil moisture data and meteorological information of the crop growth environment, and to acquire canopy multispectral images of crops collected using a multispectral sensor; The processing module is used to extract the chlorophyll absorption index of crops at the current growth stage and the vegetation index of the growth environment from the canopy multispectral image, perform nutrient diagnosis on the crop growth environment based on the chlorophyll absorption index and the vegetation index, obtain the nutrient stress status of each nutrient element, and determine the target fertilizer concentration of each nutrient element through the nutrient stress status and the nutrient distribution characteristics of the crop growth environment. The processing module is also used to determine the crop coefficient of the crop at the current growth stage, and to perform water balance on the crop growth environment based on the crop coefficient, the water distribution characteristics in the soil moisture data and the rainfall and evaporation characteristics in the meteorological information, so as to obtain the net water requirement of the crop at the current growth stage. The execution module is used to generate fertilization control commands that directly drive the integrated water and fertilizer equipment based on each target fertilizer solution concentration and the net water requirement.

9. A computer device, characterized in that, The computer device includes a memory and a processor. The memory is used to store computer programs, and the processor is used to call and run the computer programs from the memory, so that the computer device performs the artificial intelligence-based water and fertilizer control method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions or code that, when executed on a computer, cause the computer to implement the artificial intelligence-based water and fertilizer control method as described in any one of claims 1 to 7.

Images