Soil multi-layer sensing real-time acquisition and demand-supply ratio model water and fertilizer regulation method

Abstract

The invention relates to the technical field of water and fertilizer regulation and control, and discloses a soil multi-layer sensing real-time acquisition and demand-to-supply ratio model water and fertilizer regulation and control method, which comprises the following steps: firstly acquiring basic information of soil, arranging a multi-parameter sensing module to acquire moisture, conductivity, temperature, crop stem flow rate and soil texture data, encrypting effective data, and transmitting the encrypted data to an edge end. And the edge end uses a public key to decrypt and establish data, time and position mapping, and combines the preprocessed NDVI data of the unmanned aerial vehicle to fuse multi-modal data according to a matching threshold. And calculating water and fertilizer demand parameters, if the water and fertilizer demand parameters do not accord with a collaborative threshold value, adjusting a texture adaptation coefficient to recalculate, determining priorities according to regulation and control instruction types according to crop growth stages, encrypting and transmitting to an execution terminal. After execution, a dynamic data fixed tracking parameter fine-tuning acquisition frequency is obtained, data association storage encryption is carried out, and a frequency adjustment mechanism for distributed key management, abnormal fault alarm and rainfall and stem flow rate triggering is further included.

Description

Technical Field

[0001] This invention relates to the field of water and fertilizer regulation technology, and more specifically, to a method for real-time data acquisition and demand-supply ratio model-based water and fertilizer regulation using multi-layer soil sensing. Background Technology

[0002] Precise regulation of soil water and fertilizer is a core technological support for modern agriculture in ensuring food security, improving resource utilization efficiency, and mitigating environmental pollution. The dynamic demand for soil moisture and nutrients throughout the entire crop growth cycle is closely related to the physicochemical properties of soil layers (top 0-20cm, root zone 20-60cm, and deeper layers). Only by accurately capturing changes in soil parameters across multiple layers and matching them with crop supply and demand can efficient water and fertilizer utilization be achieved, avoiding resource waste and non-point source pollution in farmland. However, with the increasing demands for precision in smart agriculture, traditional water and fertilizer regulation methods based on manual experience or single-point shallow soil monitoring can no longer meet the needs of crops for dynamic perception and precise response to the soil environment, highlighting increasingly prominent technological bottlenecks.

[0003] Traditional soil water and fertilizer regulation technologies suffer from multiple deficiencies: At the parameter acquisition level, they only cover single parameters of the soil surface, lacking multi-layer monitoring of key root layers. Fixed acquisition intervals cannot cope with sudden events such as rainfall and rapid changes in salinity, and the absence of an effective mechanism for identifying and verifying abnormal data easily leads to data distortion. At the data processing and supply-demand matching level, they rely solely on soil sensor data, failing to integrate with UAV remote sensing NDVI data for correlation analysis. Furthermore, they do not consider the differences in water and fertilizer retention between different soil textures such as sandy soil and clay soil, and the use of a uniform standard results in low accuracy in supply-demand judgment. At the regulation execution and feedback level, they lack a closed-loop mechanism, making it impossible to evaluate the regulation effect and correct instructions through real-time parameters. Furthermore, fragmented data storage without secure encryption and spatiotemporal mapping makes it difficult to optimize models based on historical data. Additionally, they fail to set regulation priorities for different growth stages such as crop grain filling and seedling stages, easily leading to supply mismatches during critical periods.

[0004] Therefore, it is necessary to design a method for real-time data acquisition and water and fertilizer regulation based on multi-layer soil sensing and demand-supply ratio model, which can specifically address the problems of lack of multi-layer soil parameter acquisition, insufficient data fusion and anomaly identification capabilities, poor adaptability of demand-supply ratio calculation to soil texture, lack of closed-loop regulation feedback, and lack of data security in traditional technologies. Summary of the Invention

[0005] In view of this, the present invention proposes a method for real-time acquisition of soil multi-layer sensing and water and fertilizer regulation based on demand-supply ratio model, which aims to solve the problems of missing soil multi-layer parameter acquisition, insufficient data fusion and anomaly identification capabilities, poor adaptability of demand-supply ratio calculation to soil texture, lack of closed-loop regulation feedback, and lack of data security in traditional technologies.

[0006] In one aspect, this invention proposes a method for real-time data acquisition and water and fertilizer regulation using a multi-layer soil sensing and demand-supply ratio model, comprising: Acquire basic soil information of the monitoring area, deploy multi-parameter sensing modules at different depths based on the basic information, set an initial data acquisition interval, and collect soil moisture, soil conductivity, soil temperature and crop stem flow rate parameters at the initial interval, while simultaneously collecting soil texture type data through dielectric constant. The abnormal data pre-labeling sub-unit is used to identify out-of-range values ​​in the collected data, and a normal fluctuation threshold range is preset in advance; When there is no data outside the range, the collected data is deemed valid, and the data is packaged, encrypted, and transmitted to the edge processing terminal. When out-of-range data is found, it is determined that the collected data needs to be verified. The abnormal values ​​are marked with a verification mark, the collection interval is shortened to half of the original interval, and then the data is collected again. Obtain the encrypted raw data packet, decrypt it using the private key provided by the distributed key management system, and establish a mapping relationship between the data and time and location; By combining UAV remote sensing NDVI data correlation analysis, a data fusion matching threshold is set, and multimodal data integration is completed based on the threshold. Crop water and fertilizer demand parameters are calculated based on an integrated dataset, and the collaborative threshold of the demand parameters is preset. When the parameters meet the coordination threshold, the required parameters are deemed valid, a control command is generated, and the command is transmitted in encrypted form. When the parameter exceeds the coordination threshold, it is determined that the required parameter needs to be corrected. After adjusting the soil texture adaptation coefficient, it is recalculated, and the corrected control command is generated and executed.

[0007] Furthermore, after generating and executing the revised control instructions, the process also includes: During the execution of the modified control command, dynamic change data of soil moisture, soil conductivity, soil temperature and crop stem flow rate in the monitoring area are acquired based on the multi-parameter sensing module. Control effect tracking parameters are determined based on the acquired dynamic change data. After fine-tuning the data acquisition frequency based on the control effect tracking parameters, the final data acquisition frequency is determined. Soil and crop parameters in the monitoring area are continuously acquired based on the final data acquisition frequency. The acquired dynamic parameters are associated with the control command execution record and soil texture type data and stored in the local database. At the same time, the associated stored data is encrypted through the distributed key management module.

[0008] Furthermore, the step of determining the control effect tracking parameters based on the acquired dynamic change data, and then fine-tuning the data acquisition frequency based on the control effect tracking parameters to determine the final data acquisition frequency, includes: The soil moisture change rate, soil electrical conductivity fluctuation, soil temperature change range, and crop stem flow rate change per unit time after regulation were extracted, and these four parameters were used as the core indicators for tracking the regulation effect. The first preset moisture change rate, the second preset moisture change rate, the first preset conductivity fluctuation value, the second preset conductivity fluctuation value, the first preset temperature change range, the second preset temperature change range, the first preset stem flow change amount, and the second preset stem flow change amount are preset. At the same time, the first tracking parameter level, the second tracking parameter level, and the third tracking parameter level are set sequentially from small to large. The extracted soil moisture change rate, soil electrical conductivity fluctuation value, soil temperature change range, and crop stem flow rate change were compared with their respective preset thresholds: When the soil moisture change rate is less than or equal to the first preset moisture change rate, the soil electrical conductivity fluctuation value is less than or equal to the first preset electrical conductivity fluctuation value, the soil temperature change amplitude is less than or equal to the first preset temperature change amplitude, and the crop stem flow rate change is less than or equal to the first preset stem flow rate change, the control effect tracking parameter level is determined to be the first tracking parameter level. When the soil moisture change rate is greater than the first preset moisture change rate and less than or equal to the second preset moisture change rate, the soil electrical conductivity fluctuation value is greater than the first preset electrical conductivity fluctuation value and less than or equal to the second preset electrical conductivity fluctuation value, the soil temperature change amplitude is greater than the first preset temperature change amplitude and less than or equal to the second preset temperature change amplitude, and the crop stem flow rate change is greater than the first preset stem flow change amount and less than or equal to the second preset stem flow change amount, the control effect tracking parameter level is determined to be the second tracking parameter level. When the soil moisture change rate is greater than the second preset moisture change rate, the soil conductivity fluctuation value is greater than the second preset conductivity fluctuation value, the soil temperature change range is greater than the second preset temperature change range, and the crop stem flow rate change is greater than the second preset stem flow rate change, the control effect tracking parameter level is determined to be the third tracking parameter level. Fine-tune the data acquisition frequency based on the determined tracking parameter levels: If the level is the first tracking parameter level, the initial acquisition frequency will be reduced by 20% from the original level as the frequency to be confirmed. If the level is the second tracking parameter level, maintain the initial acquisition frequency as the frequency to be confirmed. If the level is the third tracking parameter level, the initial acquisition frequency will be increased by 20% from the original level as the frequency to be confirmed. Collect dynamic change data again to verify the integrity and stability of the data at the frequency to be confirmed. If there is no missing data and the fluctuation is within the range of normal change data, then the frequency to be confirmed will be determined as the final data collection frequency. If the data is missing or fluctuates beyond the range of changes, fine-tune it based on the tracking parameter level until a final data collection frequency that meets the requirements is determined. In determining the tracking parameter level, if any condition meets the conditions for the next level, it is directly determined to be the next tracking parameter level.

[0009] Furthermore, the step of combining UAV remote sensing NDVI data correlation analysis, setting a data fusion matching threshold, and completing multimodal data integration based on the threshold includes: The raw NDVI data is processed for cloud removal and radiometric correction, and the vegetation coverage index of the monitoring area is extracted. A first data fusion matching threshold and a second data fusion matching threshold are preset, with the first data fusion matching threshold being less than the second data fusion matching threshold; a first fusion weight and a second fusion weight are preset, with the first fusion weight being less than the second fusion weight; Calculate the degree of matching between soil moisture data, soil electrical conductivity data, soil temperature data, and NDVI data: When the matching degree is less than or equal to the first data fusion matching threshold, the data correlation is determined to be low. The first fusion weight is used to perform weighted fusion of soil sensor data and NDVI data, and the weight of soil sensor data is retained first. When the matching degree is greater than the first data fusion matching threshold and less than or equal to the second data fusion matching threshold, the data correlation is determined to be moderate, and the second fusion weight is used to perform weighted fusion of soil sensor data and NDVI data. When the matching degree is greater than the second data fusion matching threshold, the data correlation is determined to be high, and the two types of data are directly fused with equal weights. After fusion, a standardized dataset is generated, and an association index is established between the data and the land parcel identifiers in the monitoring area. This association index is used for subsequent calculation of demand parameters.

[0010] Furthermore, the calculation of crop water and fertilizer requirement parameters based on the integrated dataset, and the pre-setting of coordination thresholds for these requirement parameters, includes: Based on the integrated dataset, crop transpiration and evaporation are calculated, and nutrient demand coefficients are calculated by combining soil electrical conductivity data. A first collaboration threshold and a second collaboration threshold are preset, with the first collaboration threshold being less than the second collaboration threshold; a first texture adaptation coefficient and a second texture adaptation coefficient are preset, with the first texture adaptation coefficient being less than the second texture adaptation coefficient; The degree of synergy between crop transpiration and nutrient demand coefficient is used as an indicator of the synergy of demand parameters: When the degree of synergy is less than or equal to the first synergy threshold, it is determined that the synergy of the demand parameters is poor, and the texture adaptation coefficient needs to be adjusted. If the soil texture is sandy soil, the second texture adaptation coefficient is used to correct the nutrient demand coefficient; if it is clay soil, the first texture adaptation coefficient is used to correct the nutrient demand coefficient. When the degree of synergy is greater than the first synergy threshold and less than or equal to the second synergy threshold, the synergy of the demand parameter is determined to be moderate, the coefficient after the previous round of adjustment is maintained, and the crop transpiration evaporation is finely adjusted by ±5%. When the degree of synergy is greater than the second synergy threshold, the synergy of the demand parameters is determined to be excellent, and the crop transpiration and evaporation and the nutrient demand coefficient are directly used as the final crop water and fertilizer demand parameters. The revised demand parameters must meet the following conditions: crop transpiration and evaporation must be between the minimum and maximum crop transpiration and evaporation, and nutrient demand coefficient must be between the minimum and maximum nutrient demand coefficient. If the parameters are outside these ranges, they must be recalculated.

[0011] Furthermore, the generation and encrypted transmission of control instructions includes: Based on the final water and fertilizer demand parameters, the types of regulation instructions are classified as: water regulation instructions, nutrient regulation instructions, and comprehensive regulation instructions. Pre-set the first, second, and third command priorities in ascending order; determine the priorities based on the crop growth stage. When the crop is in the grain-filling stage, set the water regulation command priority as the first command priority and the nutrient regulation command priority as the second command priority; When the crop is in the seedling stage, set the nutrient regulation command priority as the first command priority and the water regulation command as the second command priority; When soil parameters exceed the threshold simultaneously, a comprehensive control instruction is generated with the priority of the first instruction. The control instructions are encrypted using a session key generated by a distributed key management module. After encryption, the instructions are transmitted to the water and fertilizer execution terminal via a long-distance radio gateway. The validity period of the instructions is set, and if the instructions are not executed before the validity period expires, they are re-encrypted and sent.

[0012] Furthermore, the private key decryption provided by the distributed key management establishes a mapping relationship between data and time and location, including: Assign a unique public and private key to each monitoring area; After receiving the encrypted data packet, the edge processing terminal calls the corresponding area private key stored locally to decrypt it and extract the timestamp and location coordinates from the data packet; Preset time synchronization threshold and position deviation threshold: When the time difference between the decrypted timestamp and the edge processing system is less than or equal to the time synchronization threshold, the time synchronization is considered valid; otherwise, the time is marked as abnormal, and a request is made to resend the data packet with the timestamp. When the deviation between the decrypted location coordinates and the preset monitoring area coordinates is less than or equal to the location deviation threshold, the location mapping is deemed valid; otherwise, the location is marked as abnormal, and the sensor module positioning calibration is triggered. A three-dimensional mapping table of timestamps, location coordinates, and soil parameters is created for valid data and stored in a distributed database, supporting retrieval by time and location.

[0013] Furthermore, the abnormal data pre-labeling subunit identifies out-of-range values ​​in the collected data, including: Based on the basic soil information, the conventional fluctuation threshold ranges for each parameter are preset: soil moisture fluctuation threshold range, soil electrical conductivity fluctuation threshold range, soil temperature fluctuation threshold range, and crop stem flow rate fluctuation threshold range. Each of the real-time collected parameter values ​​is evaluated individually: When the parameter value is within the normal fluctuation threshold range of the corresponding parameter, it is marked as normal data and included in the valid dataset; When a parameter value is less than the minimum value of the corresponding parameter fluctuation threshold range or greater than the maximum value of the corresponding parameter fluctuation threshold range, it is marked as out-of-range data and the abnormal type is recorded. The threshold for abnormal frequency is set to three times. If the same parameter is collected out of range three times in a row, the sensor module is judged to be faulty and a fault alarm is sent to the management platform. If there is only a single abnormality, it is marked as data to be verified and a re-collection process with a shortened collection interval is triggered.

[0014] Furthermore, the step of acquiring dynamically changing data based on the multi-parameter sensing module and storing it in a local database includes: A dual-layer storage architecture consisting of a real-time database and a historical database is adopted. The real-time database is used to store dynamically changing data within 24 hours, and the update frequency is consistent with the collection frequency; the historical database is used to archive data by day and retain a one-year historical record. Dynamically changing data includes parameters before regulation, real-time parameters during regulation, and stable parameters after regulation. When storing the data, the regulation instruction identifier and the execution terminal number are attached. Pre-set data compression threshold: When the rate of change of continuously collected parameters is less than or equal to the data compression threshold, the differential compression algorithm is used for storage, and only key data points with a rate of change greater than the data compression threshold are retained; The stored data is encrypted using a public key, and the key is updated synchronously with the distributed key management module every seven days. Historical data is backed up before each update.

[0015] Furthermore, the continuous data collection according to the final data collection frequency includes: Based on the final data acquisition frequency, the preset frequency adjustment trigger conditions are as follows: When the daily rainfall in the monitored area is greater than or equal to the threshold of the first day's rainfall, the trigger frequency is increased to 1.5 times the final data collection frequency, and the duration is 24 hours. When the rate of change of crop stem flow rate is less than or equal to the second stem flow rate change rate threshold, the trigger frequency is reduced to 0.5 times the final data acquisition frequency, and the duration is twelve hours. Among them, upward adjustment instructions for daily rainfall greater than or equal to the first daily rainfall threshold take precedence over downward adjustment instructions for stem flow rate change less than or equal to the second stem flow rate change threshold. After a frequency adjustment, a frequency change log is generated, recording the reason for the adjustment, the frequencies before and after the adjustment, and the effective time. If the adjusted frequency exceeds the preset range, the range boundary value will be taken as the final frequency. During continuous data collection, the frequency is checked for reasonableness every six hours. If the triggering condition disappears, the original final data collection frequency is restored.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. By deploying multi-parameter sensing modules at different depths, multi-level synchronous acquisition of soil moisture, electrical conductivity, temperature, and crop stem flow rate is achieved. Combined with dielectric constant, soil texture type is obtained, breaking through the limitations of traditional single-point shallow monitoring. It can more realistically reflect the vertical physicochemical properties of soil and the physiological state of crops. At the same time, the abnormal data pre-labeling and re-collection mechanism can effectively filter invalid data, ensuring the reliability of the collected data and providing high-quality data support for subsequent control decisions.

[0017] 2. The use of distributed key management for private key decryption ensures the security of data transmission and storage, avoiding the risk of information leakage; multimodal correlation analysis is performed by combining UAV remote sensing NDVI data, and data integration is achieved by setting a fusion matching threshold, so that soil physicochemical parameters and crop canopy growth information complement each other, improving the completeness of the dataset and the comprehensiveness of demand judgment.

[0018] 3. When calculating crop water and fertilizer requirements parameters based on integrated datasets, the parameters can be adapted to different soil types such as sandy soil and clay soil by pre-setting collaborative thresholds and adjusting them in combination with soil texture adaptation coefficients, thus solving the problem of "one-size-fits-all" in traditional regulation. At the same time, the correction mechanism for abnormal parameter conditions further ensures the pertinence and effectiveness of regulation instructions, which can significantly improve water and fertilizer use efficiency and reduce resource waste and environmental pollution. Attached Figure Description

[0019] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a first flowchart of the soil multilayer sensing real-time acquisition and demand-supply ratio model water and fertilizer regulation method provided in an embodiment of the present invention. Figure 2 This is the second flowchart of the soil multilayer sensing real-time acquisition and demand-supply ratio model water and fertilizer regulation method provided in the embodiments of the present invention. Detailed Implementation

[0020] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] Reference Figure 1 In some embodiments of this application, a method for real-time soil multilayer sensing and water and fertilizer regulation based on a demand-supply ratio model includes: Step S100: Obtain basic soil information of the monitoring area, deploy multi-parameter sensing modules at different depths based on the basic information, set the initial data acquisition interval, and collect soil moisture, soil conductivity, soil temperature and crop stem flow rate parameters at the initial interval, while simultaneously collecting soil texture type data through dielectric constant. Step S200: Use the abnormal data pre-labeling sub-unit to identify out-of-range values ​​in the collected data and pre-set the normal fluctuation threshold range; When there is no data outside the range, the collected data is deemed valid, and the data is packaged, encrypted, and then transmitted to the edge processing end. When out-of-range data is found, the collected data is determined to need to be verified. The outlier is marked with a verification mark, the collection interval is shortened to half of the original interval, and then the data is collected again. Step S300: Obtain the encrypted original data packet, decrypt it using the private key provided by the distributed key management, and establish a mapping relationship between the data and time and location; Step S400: Combine UAV remote sensing NDVI data correlation analysis, set data fusion matching thresholds, and complete multimodal data integration based on the thresholds; Step S500: Calculate crop water and fertilizer demand parameters based on the integrated dataset, and pre-set the coordination threshold of the demand parameters; When the parameters meet the collaborative threshold, the required parameters are deemed valid, control instructions are generated and transmitted in encrypted form. When the parameters exceed the coordination threshold, it is determined that the required parameters need to be corrected. The soil texture adaptation coefficient is adjusted and recalculated, and the corrected control command is generated and executed.

[0022] The above embodiments, by deploying multi-parameter sensing modules at different depths, achieve multi-level synchronous acquisition of soil moisture, electrical conductivity, temperature, and crop stem flow rate. Combined with dielectric constant to obtain soil texture type, this overcomes the limitations of traditional single-point shallow monitoring and can more realistically reflect the longitudinal physicochemical properties of the soil and the physiological state of the crop. At the same time, the abnormal data pre-labeling and re-collection mechanism can effectively filter invalid data, ensuring the reliability of the collected data and providing high-quality data support for subsequent regulatory decisions.

[0023] The above embodiments employ a distributed key management private key decryption method to ensure the security of data transmission and storage and avoid the risk of information leakage. By combining UAV remote sensing NDVI data for multimodal correlation analysis and setting a fusion matching threshold to achieve data integration, soil physicochemical parameters and crop canopy growth information complement each other, thereby improving the completeness of the dataset and the comprehensiveness of demand judgment.

[0024] When calculating crop water and fertilizer requirements parameters based on the integrated dataset, the above embodiments adjust the parameters by setting a collaborative threshold and combining it with the soil texture adaptation coefficient, so that the requirements parameters can be adapted to different soil types such as sandy soil and clay soil, thus solving the problem of "one-size-fits-all" in traditional regulation. At the same time, the correction mechanism for abnormal parameter conditions further ensures the pertinence and effectiveness of the regulation instructions, which can significantly improve water and fertilizer utilization efficiency and reduce resource waste and environmental pollution.

[0025] Reference Figure 2 In some embodiments of this application, after generating and executing the modified control command, the method further includes: Step S600: During the execution of the modified control command, dynamic change data of soil moisture, soil conductivity, soil temperature and crop stem flow rate in the monitoring area are acquired based on the multi-parameter sensing module. The control effect tracking parameters are determined based on the acquired dynamic change data. After fine-tuning the data acquisition frequency based on the control effect tracking parameters, the final data acquisition frequency is determined. Soil and crop parameters in the monitoring area are continuously acquired based on the final data acquisition frequency. The acquired dynamic parameters are associated with the control command execution record and soil texture type data and stored in the local database. At the same time, the associated stored data is encrypted through the distributed key management module.

[0026] Specifically, the control effect tracking parameters are determined based on the acquired dynamic change data. The final data acquisition frequency is then determined after fine-tuning the data acquisition frequency based on these parameters, including: The soil moisture change rate, soil electrical conductivity fluctuation, soil temperature change range, and crop stem flow rate change per unit time after regulation were extracted, and these four parameters were used as the core indicators for tracking the regulation effect. The first preset moisture change rate, the second preset moisture change rate, the first preset conductivity fluctuation value, the second preset conductivity fluctuation value, the first preset temperature change range, the second preset temperature change range, the first preset stem flow change amount, and the second preset stem flow change amount are preset. At the same time, the first tracking parameter level, the second tracking parameter level, and the third tracking parameter level are set sequentially from small to large. The extracted soil moisture change rate, soil electrical conductivity fluctuation value, soil temperature change range, and crop stem flow rate change were compared with their respective preset thresholds: When the soil moisture change rate is less than or equal to the first preset moisture change rate, the soil electrical conductivity fluctuation value is less than or equal to the first preset electrical conductivity fluctuation value, the soil temperature change amplitude is less than or equal to the first preset temperature change amplitude, and the crop stem flow rate change is less than or equal to the first preset stem flow rate change, the control effect tracking parameter level is determined to be the first tracking parameter level. When the soil moisture change rate is greater than the first preset moisture change rate and less than or equal to the second preset moisture change rate, the soil electrical conductivity fluctuation value is greater than the first preset electrical conductivity fluctuation value and less than or equal to the second preset electrical conductivity fluctuation value, the soil temperature change amplitude is greater than the first preset temperature change amplitude and less than or equal to the second preset temperature change amplitude, and the crop stem flow rate change is greater than the first preset stem flow change amount and less than or equal to the second preset stem flow change amount, the control effect tracking parameter level is determined to be the second tracking parameter level. When the soil moisture change rate is greater than the second preset moisture change rate, the soil conductivity fluctuation value is greater than the second preset conductivity fluctuation value, the soil temperature change range is greater than the second preset temperature change range, and the crop stem flow rate change is greater than the second preset stem flow rate change, the control effect tracking parameter level is determined to be the third tracking parameter level. Fine-tune the data acquisition frequency based on the determined tracking parameter levels: If the level is the first tracking parameter level, the initial acquisition frequency will be reduced by 20% from the original level as the frequency to be confirmed. If the level is the second tracking parameter level, maintain the initial acquisition frequency as the frequency to be confirmed. If the level is the third tracking parameter level, the initial acquisition frequency will be increased by 20% from the original level as the frequency to be confirmed; Collect dynamic change data again to verify the integrity and stability of the data at the frequency to be confirmed. If there is no missing data and the fluctuation is within the range of normal change data, then the frequency to be confirmed will be determined as the final data collection frequency. If the data is missing or fluctuates beyond the range of changes, fine-tune it based on the tracking parameter level until a final data collection frequency that meets the requirements is determined. In determining the tracking parameter level, if any condition meets the conditions for the next level, it is directly determined to be the next tracking parameter level.

[0027] Specifically, when executing the revised control instructions, dynamic data on soil moisture, electrical conductivity, temperature and crop stem flow rate are collected continuously for 8 hours at a rate of 1 hour per session by multi-parameter sensing modules deployed in the surface layer (0-20cm), root layer (20-60cm), and deep layer (60-100cm) (2 nodes per layer).

[0028] Specifically, the core tracking indicators are the soil moisture change rate, electrical conductivity fluctuation, temperature change amplitude, and stem flow rate change within one hour after regulation. Thresholds are set according to the soil types: sandy soil, clay soil, and loam (e.g., the first preset moisture change rate for sandy soil is ≤0.8% / h, and the second is ≤1.5% / h; for clay soil, the first is ≤0.5% / h, and the second is ≤1.0% / h; for loam, the first is ≤0.6% / h, and the second is ≤1.2% / h; other indicator thresholds are set according to soil texture). The level determination follows the principle of "upgrading if any one of the indicators reaches the next level", and is divided into three levels (level 1: all four indicators are ≤ the first threshold; level 2: any one or all of them are between the first and second thresholds; level 3: any one indicator is > the second threshold). The corresponding initial frequency (1 time / 2 hours) is reduced by 20% (1 time / 2.5 hours), maintained, and increased by 20% (1 time / 1.6 hours) as the frequency to be confirmed. After 4 hours of data collection and verification (missing rate ≤5% and fluctuation does not exceed the second threshold), the final frequency is determined. If the conditions are not met, it is fine-tuned again.

[0029] The above embodiments comprehensively reflect the control effect by extracting four core indicators, including soil moisture change rate and electrical conductivity fluctuation value. Combined with the tiered preset threshold and the judgment rule of "upgrading if any one indicator meets the standard", it can accurately capture abnormal fluctuations of parameters to avoid the risk of missed judgments. It can also fine-tune the collection frequency according to the level (lowering by 20% for the first level to save resources and increasing by 20% for the third level to ensure data integrity). At the same time, a closed loop is formed through data integrity and stability verification, which ultimately achieves dynamic adaptation between collection frequency and control effect. This ensures that key data are not missing and avoids the waste of resources caused by ineffective collection, thereby improving the accuracy and efficiency of soil and crop parameter collection.

[0030] Specifically, by combining UAV remote sensing NDVI data correlation analysis, a data fusion matching threshold is set, and multimodal data integration is completed based on the threshold, including: The raw NDVI data is processed for cloud removal and radiometric correction, and the vegetation coverage index of the monitoring area is extracted. A first data fusion matching threshold and a second data fusion matching threshold are preset, with the first data fusion matching threshold being less than the second data fusion matching threshold; a first fusion weight and a second fusion weight are preset, with the first fusion weight being less than the second fusion weight; Calculate the degree of matching between soil moisture data, soil electrical conductivity data, soil temperature data, and NDVI data: When the matching degree is less than or equal to the first data fusion matching threshold, the data correlation is determined to be low. The first fusion weight is used to perform weighted fusion of soil sensor data and NDVI data, and the weight of soil sensor data is retained first. When the matching degree is greater than the first data fusion matching threshold and less than or equal to the second data fusion matching threshold, the data correlation is determined to be moderate, and the second fusion weight is used to perform weighted fusion of soil sensor data and NDVI data. When the matching degree is greater than the second data fusion matching threshold, the data correlation is determined to be high, and the two types of data are directly fused with equal weights. After fusion, a standardized dataset is generated, and an association index is established between the data and the land parcel identifiers in the monitoring area. The association index is used for subsequent calculation of required parameters.

[0031] Specifically, after declouding and radiometric correction of the raw NDVI data using ENVI software, the vegetation cover index of the monitoring area is extracted according to the formula: "Vegetation Cover Index = (NDVI value in the monitoring area - minimum NDVI value in the monitoring area) ÷ (maximum NDVI value in the monitoring area - minimum NDVI value in the monitoring area)". The first data fusion matching threshold is preset to 0.3, the second data fusion matching threshold is preset to 0.7, the first fusion weight (soil sensor data: NDVI data) is 0.7:0.3, and the second fusion weight is 0.6:0.4. The Peel method is used. The correlation coefficient is used to calculate the matching degree between soil moisture, electrical conductivity, and temperature data and NDVI data. When the matching degree is ≤0.3, the correlation is considered low, and the data is fused with a weight of 0.7:0.3, with priority given to retaining the weight of soil sensor data. When the matching degree is 0.3 < matching degree ≤0.7, the correlation is considered medium, and the data is fused with a weight of 0.6:0.4. When the matching degree is >0.7, the correlation is considered high, and the data is fused with an equal weight of 0.5:0.5. After fusion, a standardized dataset with a uniform format is generated, and an association index is established between the data and the unique code of the monitored plot (such as "plot-001") for subsequent calculation of crop water and fertilizer requirements parameters.

[0032] The above embodiments ensure the accuracy of the vegetation coverage index by performing cloud removal and radiometric correction on the original NDVI data. Combined with differentiated settings of dual data fusion matching thresholds (first threshold < second threshold) and dual fusion weights (first weight < second weight), the fusion strategy can be dynamically adjusted based on the matching degree between soil moisture, electrical conductivity, temperature data, and NDVI data. When the correlation is low, soil sensor data weights are prioritized to ensure accurate micro-parameters; when the correlation is medium, the second weight balances the two types of data, taking into account both micro and macro information; and when the correlation is high, equal-weighted fusion maximizes the synergistic value of the data. Finally, a standardized dataset with plot identification and correlation index is generated. This achieves the complementary advantages of soil sensor data (precisely reflecting local water and fertilizer conditions) and NDVI data (macroscopically reflecting vegetation growth status), and provides accurate and correlated high-quality data support for subsequent crop water and fertilizer demand parameter calculations, effectively improving the comprehensiveness and accuracy of demand judgment.

[0033] Specifically, crop water and fertilizer requirement parameters are calculated based on an integrated dataset, and pre-set coordination thresholds for these requirements parameters, including: Based on the integrated dataset, crop transpiration and evaporation are calculated, and nutrient demand coefficients are calculated by combining soil electrical conductivity data. A first collaboration threshold and a second collaboration threshold are preset, with the first collaboration threshold being less than the second collaboration threshold; a first texture adaptation coefficient and a second texture adaptation coefficient are preset, with the first texture adaptation coefficient being less than the second texture adaptation coefficient; The degree of synergy between crop transpiration and nutrient demand coefficient is used as an indicator of the synergy of demand parameters: When the degree of synergy is less than or equal to the first synergy threshold, it is determined that the synergy of the demand parameters is poor, and the texture adaptation coefficient needs to be adjusted. If the soil texture is sandy soil, the second texture adaptation coefficient is used to correct the nutrient demand coefficient; if it is clay soil, the first texture adaptation coefficient is used to correct the nutrient demand coefficient. When the degree of synergy is greater than the first synergy threshold and less than or equal to the second synergy threshold, the synergy of the demand parameter is determined to be moderate, the coefficient after the previous round of adjustment is maintained, and the crop transpiration evaporation is finely adjusted by ±5%. When the degree of synergy is greater than the second synergy threshold, the synergy of the demand parameters is determined to be excellent, and the crop transpiration and evaporation and the nutrient demand coefficient are directly used as the final crop water and fertilizer demand parameters. The revised demand parameters must meet the following conditions: crop transpiration and evaporation must be between the minimum and maximum crop transpiration and evaporation, and nutrient demand coefficient must be between the minimum and maximum nutrient demand coefficient. If the parameters are outside these ranges, they must be recalculated.

[0034] Specifically, based on the integrated dataset, crop transpiration and evaporation are calculated using the Penman-Montes formula. Combined with soil electrical conductivity data, the nutrient requirement coefficient is calculated using the formula "nutrient requirement coefficient = soil electrical conductivity × 0.35". A first synergy threshold of 0.4 and a second synergy threshold of 0.7 (first synergy threshold < second synergy threshold) are pre-set, along with a first texture fit coefficient of 0.8 and a second texture fit coefficient of 1.2 (first texture fit coefficient < second texture fit coefficient). The synergy between crop transpiration and evaporation and the nutrient requirement coefficient is used as an indicator: when the synergy is ≤ 0.4 (…),… (Poor synergy) Use a texture compatibility coefficient of 1.2 for sandy soil, 0.8 for clay soil, and 1.0 for loam (default median value) to correct the nutrient requirement coefficient; when 0.4 < synergy ≤ 0.7 (medium synergy), maintain the previous coefficient and finely adjust the transpiration evaporation by ±5%; when synergy > 0.7 (excellent synergy), directly use both as the final parameters, and after correction, the crop transpiration evaporation must be between 2 mm / d (minimum) and 8 mm / d (maximum) and the nutrient requirement coefficient must be between 0.5 (minimum) and 1.8 (maximum). If they exceed these limits, recalculate.

[0035] Specifically, the calculation method using the Penman-Montes formula is as follows: Actual crop evapotranspiration (ETc) = Reference crop evapotranspiration (ET0) × Crop coefficient (Kc) The formula for calculating the reference crop evapotranspiration (ET0) is: ET0 = [0.408 × Δ × (Rn - G) + γ × (900 / (T + 273)) × u² × (es - ea)] / [Δ + γ × (1 + 0.34 × u²)] The parameters are as follows: Δ: Slope of the curve relating saturated vapor pressure to temperature (kPa / ℃); Rn: Net radiation received by the crop canopy surface (MJ / (m²・d)); G: Soil heat flux density (MJ / (m²・d), usually taken as 0.1×Rn during the day and 0.5×Rn at night); γ: Wet and dry surface constant (kPa / ℃, usually taken as 0.665kPa / ℃); T: Daily average air temperature (°C) at a height of 2m; u2: Wind speed at a height of 2m (m / s); es: Saturated vapor pressure (kPa, calculated from air temperature T); ea: Actual water vapor pressure (kPa, calculated from relative humidity of the air); Kc: Crop coefficient (unitless, determined according to crop type and growth stage, e.g., Kc is about 0.8 during the jointing stage of wheat and about 1.1 during the grain-filling stage).

[0036] The above embodiments calculate crop transpiration (reflecting water demand) and nutrient demand coefficients (combining soil sensor and NDVI data) based on an integrated dataset (fusion of soil sensor and NDVI data). By setting a first synergy threshold (low synergy standard), a second synergy threshold (high synergy standard), and first / second texture adaptation coefficients for different soil textures, the demand parameters can be dynamically optimized based on the synergy between the two. When the synergy is poor, the nutrient demand coefficients are precisely corrected for sandy soil (using the second coefficient) and clay soil (using the first coefficient). When the synergy is moderate, the transpiration is fine-tuned while maintaining the adaptation coefficients. When the synergy is excellent, the parameters are directly used. At the same time, minimum / maximum threshold constraints ensure that the corrected parameters conform to the crop growth pattern. This achieves the matching of water and fertilizer demand parameters with soil texture (sandy soil has weak water and fertilizer retention, clay soil has strong water and fertilizer retention) and avoids parameter disconnect or abnormality. Finally, accurate and reasonable crop water and fertilizer demand parameters are output, providing a scientific basis for the generation of subsequent water and fertilizer regulation instructions, effectively improving the matching degree between water and fertilizer supply and crop demand, and reducing the problems of water and fertilizer waste or insufficient supply.

[0037] Specifically, generating control commands and encrypting their transmission includes: Based on the final water and fertilizer demand parameters, the types of regulation instructions are classified as: water regulation instructions, nutrient regulation instructions, and comprehensive regulation instructions. Pre-set the first, second, and third command priorities in ascending order; determine the priorities based on the crop growth stage. When the crop is in the grain-filling stage, set the water regulation command priority as the first command priority and the nutrient regulation command priority as the second command priority; When the crop is in the seedling stage, set the nutrient regulation command priority as the first command priority and the water regulation command as the second command priority; When soil parameters exceed the threshold simultaneously, a comprehensive control instruction is generated with the priority of the first instruction. The control instructions are encrypted using a session key generated by a distributed key management module. After encryption, the instructions are transmitted to the water and fertilizer execution terminal via a long-distance radio gateway. The validity period of the instructions is set, and if the instructions are not executed before the validity period expires, they are re-encrypted and sent.

[0038] The above embodiments, by classifying control commands into three categories—water, nutrients, and comprehensive—based on final water and fertilizer demand parameters, can accurately match different water and fertilizer deficiency scenarios and avoid control redundancy. Furthermore, by combining differentiated priority settings for crop growth stages—prioritizing water supply during the grain-filling stage (critical water requirement period) and nutrient supply during the seedling stage (basic growth period)—and setting comprehensive commands when soil parameters simultaneously exceed thresholds as the highest priority, the embodiments ensure that control priorities align with crop growth patterns and the urgency of actual needs. Simultaneously, the embodiments employ session key encryption using distributed key management and transmission via long-distance radio gateways, ensuring both the security (anti-tampering and anti-leakage) and field coverage of command transmission. By setting command validity periods and retransmission mechanisms for expired commands, the embodiments prevent control failure due to terminal delays or malfunctions. Ultimately, the embodiments achieve precise, scenario-based, secure, and efficient execution of water and fertilizer control commands, providing a reliable guarantee for water and fertilizer supply throughout the entire crop growth period.

[0039] Specifically, distributed key management provides private key decryption and establishes a mapping relationship between data and time and location, including: Assign a unique public and private key to each monitoring area; After receiving the encrypted data packet, the edge processing terminal calls the corresponding area private key stored locally to decrypt it and extract the timestamp and location coordinates from the data packet; Preset time synchronization threshold and position deviation threshold: When the time difference between the decrypted timestamp and the edge processing system is less than or equal to the time synchronization threshold, the time synchronization is considered valid; otherwise, the time is marked as abnormal, and a request is made to resend the data packet with the timestamp. When the deviation between the decrypted location coordinates and the preset monitoring area coordinates is less than or equal to the location deviation threshold, the location mapping is deemed valid; otherwise, the location is marked as abnormal, and the sensor module positioning calibration is triggered. A three-dimensional mapping table of timestamps, location coordinates, and soil parameters is created for valid data and stored in a distributed database, supporting retrieval by time and location.

[0040] Specifically, a unique 2048-bit RSA public key and a 2048-bit RSA private key are assigned to each monitoring area. After receiving the encrypted data packet, the edge processing terminal calls the corresponding 2048-bit RSA private key stored locally to decrypt it and extract the millisecond-level timestamp and latitude / longitude coordinates from the data packet. A time synchronization threshold of 3 seconds and a position deviation threshold of 5 meters are preset: when the time difference between the decrypted timestamp and the edge processing terminal system time is ≤3 seconds, the time synchronization is considered valid; otherwise, the time is marked as abnormal and a request is made to resend the data packet with the timestamp. When the decrypted location coordinates deviate from the preset monitoring area coordinates by ≤5 meters, the location mapping is deemed valid; otherwise, the location is marked as abnormal and the sensor module's positioning calibration is triggered. A three-dimensional mapping table of millisecond-level timestamps, latitude and longitude coordinates, and soil parameters is established for valid data and stored in the HBase distributed database, supporting retrieval by time range (e.g., "2025-08-29 08:00-12:00") and location range (e.g., "30°20′-30°25′ N, 120°10′-120°15′ E").

[0041] The above embodiments assign a unique public and private key to each monitoring area. The edge processing terminal uses the corresponding private key to decrypt the data, ensuring exclusive security of data decryption and preventing cross-regional data leakage. By setting time synchronization thresholds (e.g., 3 seconds) and location deviation thresholds (e.g., 5 meters) for verification, abnormal data with asynchronous time and misaligned locations can be filtered out, ensuring accurate correlation between data and spatiotemporal information. A three-dimensional mapping table of timestamps, location coordinates, and soil parameters is established and stored in a distributed database, which not only realizes structured data management but also supports fast retrieval by time and location. This provides a secure, accurate, and easily accessible data foundation for subsequent precise location of monitoring areas, tracking of spatiotemporal changes in data, and generation of targeted water and fertilizer control instructions, effectively improving the security, accuracy, and practicality of data management.

[0042] Specifically, the abnormal data pre-labeling sub-unit identifies out-of-range values ​​in the collected data, including: Based on basic soil information, the conventional fluctuation threshold ranges for each parameter are preset: soil moisture fluctuation threshold range, soil electrical conductivity fluctuation threshold range, soil temperature fluctuation threshold range, and crop stem flow rate fluctuation threshold range. Each of the real-time collected parameter values ​​is evaluated individually: When the parameter value is within the normal fluctuation threshold range of the corresponding parameter, it is marked as normal data and included in the valid dataset; When a parameter value is less than the minimum value of the corresponding parameter fluctuation threshold range or greater than the maximum value of the corresponding parameter fluctuation threshold range, it is marked as out-of-range data and the abnormal type is recorded. The threshold for abnormal frequency is set to three times. If the same parameter is collected out of range three times in a row, the sensor module is judged to be faulty and a fault alarm is sent to the management platform. If there is only a single abnormality, it is marked as data to be verified and a re-collection process with a shortened collection interval is triggered.

[0043] Specifically, based on basic soil information (applicable to field crops such as corn and wheat), pre-set the normal fluctuation threshold ranges for each parameter: soil moisture fluctuation threshold range is 10%–30% (volume water content), soil electrical conductivity fluctuation threshold range is 0.2–2.5 mS / cm, soil temperature fluctuation threshold range is 5–35℃, and crop stem flow rate fluctuation threshold range is 0.01–0.15 mL / (h・cm²). Each parameter value collected in real time is evaluated, and those falling within the corresponding range are marked as normal data and included in the valid dataset. Data that is less than the minimum value or greater than the maximum value is marked as out-of-range data and the anomaly type is recorded (e.g., too low moisture or too high conductivity). The pre-set threshold for the frequency of anomalies is three times. If the same parameter is collected three times in a row and the data is out of range (e.g., soil temperature is <5℃ or >35℃ for three consecutive times), the sensor module is judged to be faulty and a fault alarm is sent to the management platform. If there is only a single anomaly (e.g., soil moisture = 8%), it is marked as data to be verified and a re-collection process is triggered to shorten the original collection interval (e.g., 1 time / 2 hours) to half (1 time / 1 hour).

[0044] The above embodiments, by combining basic soil information to preset the conventional fluctuation threshold ranges of soil moisture, electrical conductivity, temperature, and crop stem flow rate, provide a precise basis for judging the validity of data. They can clearly distinguish between normal data (included in the valid dataset) and out-of-range data (recorded as anomaly type), avoiding invalid data from interfering with subsequent analysis. At the same time, using three times as the anomaly frequency threshold, the system can promptly determine the sensor module failure and send an alarm when the same parameter exceeds the range three times in a row, ensuring rapid troubleshooting of equipment problems. In the case of only a single anomaly, the system can mark the data to be verified and trigger a re-collection process with a shortened collection interval, reducing data loss caused by occasional anomalies. Ultimately, this significantly improves the validity, completeness, and reliability of the collected data, laying a high-quality data foundation for subsequent accurate calculation of crop water and fertilizer requirements and generation of control instructions.

[0045] Specifically, dynamically changing data is acquired based on a multi-parameter sensing module and stored in a local database, including: A dual-layer storage architecture consisting of a real-time database and a historical database is adopted. The real-time database is used to store dynamically changing data within 24 hours, with an update frequency consistent with the data collection frequency; the historical database is used to archive data by day and retain a year's worth of historical records. Dynamically changing data includes parameters before regulation, real-time parameters during regulation, and stable parameters after regulation. When storing the data, the regulation instruction identifier and the execution terminal number are attached. Pre-set data compression threshold: When the rate of change of continuously collected parameters is less than or equal to the data compression threshold, the differential compression algorithm is used for storage, and only key data points with a rate of change greater than the data compression threshold are retained; The stored data is encrypted using a public key, and the key is updated synchronously with the distributed key management module every seven days. Historical data is backed up before each update.

[0046] Specifically, a two-tiered storage architecture is adopted, consisting of a real-time database (storing dynamically changing data within 24 hours, with an update frequency consistent with the collection frequency; for example, if the collection frequency is once every 2 hours, the real-time database is updated every 2 hours) and a historical database (archiving data daily and retaining 1 year of historical records). The dynamically changing data includes parameters before regulation, real-time parameters during regulation, and stable parameters after regulation. When storing the data, a 16-bit coded regulation instruction identifier (e.g., "CMD-20250829001") and an 8-digit execution terminal number (e.g., "TERM-00000001") are attached. The data compression threshold is preset to 0.5%. When the continuously collected parameter change rate is ≤0.5%, the differential compression algorithm is used for storage, retaining only key data points with a change rate >0.5%. The stored data is encrypted with a 2048-bit RSA public key, and the key is updated synchronously with the distributed key management module every 7 days. Before the update, a full backup of the historical data is performed through an off-site backup server.

[0047] Specifically, data is continuously collected based on the final data collection frequency, including: Based on the final data acquisition frequency, the preset frequency adjustment trigger conditions are as follows: When the daily rainfall in the monitored area is greater than or equal to the threshold of the first day's rainfall, the trigger frequency is increased to 1.5 times the final data collection frequency, and the duration is 24 hours. When the rate of change of crop stem flow rate is less than or equal to the second stem flow rate change rate threshold, the trigger frequency is reduced to 0.5 times the final data acquisition frequency, and the duration is twelve hours. Among them, upward adjustment instructions for daily rainfall greater than or equal to the first daily rainfall threshold take precedence over downward adjustment instructions for stem flow rate change less than or equal to the second stem flow rate change threshold. After a frequency adjustment, a frequency change log is generated, recording the reason for the adjustment, the frequencies before and after the adjustment, and the effective time. If the adjusted frequency exceeds the preset range, the range boundary value will be taken as the final frequency. During continuous data collection, the frequency is checked for reasonableness every six hours. If the triggering condition disappears, the original final data collection frequency is restored.

[0048] The above-described implementation process uses dynamic conditions to trigger frequency increases (to 1.5 times the original frequency for 24 hours) based on preset daily rainfall (e.g., a first-day rainfall threshold of 50 mm) and frequency decreases (to 0.5 times the original frequency for 12 hours) based on crop stem flow rate change rate (e.g., a second threshold of 0.1% / h). This allows for both increasing the sampling frequency to capture sudden changes in soil parameters during critical environmental changes such as rainfall and reducing the frequency to minimize resource waste during periods of calm crop physiological activity. Furthermore, the rainfall-based frequency increase instruction prioritizes data integrity for critical scenarios. Simultaneously, frequency change logs ensure traceability, frequency range boundaries prevent anomalies, and the original frequency is checked and restored every 6 hours to ensure adaptability. Ultimately, this achieves precise matching between sampling frequency and environmental and crop conditions, optimizing resource consumption while ensuring data timeliness and integrity, and providing stable and reliable parameter support for subsequent water and fertilizer management.

[0049] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. 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 goods 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.

[0050] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0051] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0052] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0053] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for real-time soil multilayer sensing and water and fertilizer regulation based on a demand-supply ratio model, characterized in that, include: Acquire basic soil information of the monitoring area, deploy multi-parameter sensing modules at different depths based on the basic information, set an initial data acquisition interval, and collect soil moisture, soil conductivity, soil temperature and crop stem flow rate parameters at the initial interval, while simultaneously collecting soil texture type data through dielectric constant. The abnormal data pre-labeling sub-unit is used to identify out-of-range values ​​in the collected data, and a normal fluctuation threshold range is preset in advance; When there is no data outside the range, the collected data is deemed valid, and the data is packaged, encrypted, and transmitted to the edge processing terminal. When out-of-range data is found, it is determined that the collected data needs to be verified. The abnormal values ​​are marked with a verification mark, the collection interval is shortened to half of the original interval, and then the data is collected again. Obtain the encrypted raw data packet, decrypt it using the private key provided by the distributed key management system, and establish a mapping relationship between the data and time and location; By combining UAV remote sensing NDVI data correlation analysis, a data fusion matching threshold is set, and multimodal data integration is completed based on the threshold. Crop water and fertilizer demand parameters are calculated based on an integrated dataset, and the collaborative threshold of the demand parameters is preset. When the parameters meet the coordination threshold, the required parameters are deemed valid, a control command is generated, and the command is transmitted in encrypted form. When the parameter exceeds the coordination threshold, it is determined that the required parameter needs to be corrected. After adjusting the soil texture adaptation coefficient, it is recalculated, and the corrected control command is generated and executed.

2. The method for real-time soil multi-layer sensing and water and fertilizer regulation based on demand-supply ratio model according to claim 1, characterized in that, After generating and executing the revised control instructions, the following is also included: During the execution of the modified control command, dynamic change data of soil moisture, soil conductivity, soil temperature and crop stem flow rate in the monitoring area are acquired based on the multi-parameter sensing module. Control effect tracking parameters are determined based on the acquired dynamic change data. After fine-tuning the data acquisition frequency based on the control effect tracking parameters, the final data acquisition frequency is determined. Soil and crop parameters in the monitoring area are continuously acquired based on the final data acquisition frequency. The acquired dynamic parameters are associated with the control command execution record and soil texture type data and stored in the local database. At the same time, the associated stored data is encrypted through the distributed key management module.

3. The method for real-time soil multilayer sensing and water and fertilizer regulation based on demand-supply ratio model according to claim 2, characterized in that, The process of determining the control effect tracking parameters based on the acquired dynamic change data, and then fine-tuning the data acquisition frequency based on the control effect tracking parameters to determine the final data acquisition frequency, includes: The soil moisture change rate, soil electrical conductivity fluctuation, soil temperature change range, and crop stem flow rate change per unit time after regulation were extracted, and these four parameters were used as the core indicators for tracking the regulation effect. The first preset moisture change rate, the second preset moisture change rate, the first preset conductivity fluctuation value, the second preset conductivity fluctuation value, the first preset temperature change range, the second preset temperature change range, the first preset stem flow change amount, and the second preset stem flow change amount are preset. At the same time, the first tracking parameter level, the second tracking parameter level, and the third tracking parameter level are set sequentially from small to large. The extracted soil moisture change rate, soil electrical conductivity fluctuation value, soil temperature change range, and crop stem flow rate change were compared with their respective preset thresholds: When the soil moisture change rate is less than or equal to the first preset moisture change rate, the soil electrical conductivity fluctuation value is less than or equal to the first preset electrical conductivity fluctuation value, the soil temperature change amplitude is less than or equal to the first preset temperature change amplitude, and the crop stem flow rate change is less than or equal to the first preset stem flow rate change, the control effect tracking parameter level is determined to be the first tracking parameter level. When the soil moisture change rate is greater than the first preset moisture change rate and less than or equal to the second preset moisture change rate, the soil electrical conductivity fluctuation value is greater than the first preset electrical conductivity fluctuation value and less than or equal to the second preset electrical conductivity fluctuation value, the soil temperature change amplitude is greater than the first preset temperature change amplitude and less than or equal to the second preset temperature change amplitude, and the crop stem flow rate change is greater than the first preset stem flow change amount and less than or equal to the second preset stem flow change amount, the control effect tracking parameter level is determined to be the second tracking parameter level. When the soil moisture change rate is greater than the second preset moisture change rate, the soil conductivity fluctuation value is greater than the second preset conductivity fluctuation value, the soil temperature change range is greater than the second preset temperature change range, and the crop stem flow rate change is greater than the second preset stem flow rate change, the control effect tracking parameter level is determined to be the third tracking parameter level. Fine-tune the data acquisition frequency based on the determined tracking parameter levels: If the level is the first tracking parameter level, the initial acquisition frequency will be reduced by 20% from the original level as the frequency to be confirmed. If the level is the second tracking parameter level, maintain the initial acquisition frequency as the frequency to be confirmed. If the level is the third tracking parameter level, the initial acquisition frequency will be increased by 20% from the original level as the frequency to be confirmed; Collect dynamic change data again to verify the integrity and stability of the data at the frequency to be confirmed. If there is no missing data and the fluctuation is within the range of normal change data, then the frequency to be confirmed will be determined as the final data collection frequency. If the data is missing or fluctuates beyond the range of changes, fine-tune it based on the tracking parameter level until a final data collection frequency that meets the requirements is determined. In determining the tracking parameter level, if any condition meets the conditions for the next level, it is directly determined to be the next tracking parameter level.

4. The method for real-time data acquisition and water and fertilizer regulation using a multi-layer soil sensing and demand-supply model according to claim 3, characterized in that, The method involves combining UAV remote sensing NDVI data correlation analysis, setting a data fusion matching threshold, and completing multimodal data integration based on the threshold, including: The raw NDVI data is processed for cloud removal and radiometric correction, and the vegetation coverage index of the monitoring area is extracted. A first data fusion matching threshold and a second data fusion matching threshold are preset, wherein the first data fusion matching threshold is less than the second data fusion matching threshold; a first fusion weight and a second fusion weight are preset, wherein the first fusion weight is less than the second fusion weight; Calculate the degree of matching between soil moisture data, soil electrical conductivity data, soil temperature data, and NDVI data: When the matching degree is less than or equal to the first data fusion matching threshold, the data correlation is determined to be low. The first fusion weight is used to perform weighted fusion of soil sensor data and NDVI data, and the weight of soil sensor data is retained first. When the matching degree is greater than the first data fusion matching threshold and less than or equal to the second data fusion matching threshold, the data correlation is determined to be moderate, and the second fusion weight is used to perform weighted fusion of soil sensor data and NDVI data. When the matching degree is greater than the second data fusion matching threshold, the data correlation is determined to be high, and the two types of data are directly fused with equal weights. After fusion, a standardized dataset is generated, and an association index is established between the data and the land parcel identifiers in the monitoring area. This association index is used for subsequent calculation of demand parameters.

5. The method for real-time soil multi-layer sensing and water and fertilizer regulation based on demand-supply ratio model according to claim 4, characterized in that, The calculation of crop water and fertilizer demand parameters based on the integrated dataset, and the pre-setting of collaborative thresholds for the demand parameters, include: Based on the integrated dataset, crop transpiration and evaporation are calculated, and nutrient demand coefficients are calculated by combining soil electrical conductivity data. A first collaboration threshold and a second collaboration threshold are preset, with the first collaboration threshold being less than the second collaboration threshold; a first texture adaptation coefficient and a second texture adaptation coefficient are preset, with the first texture adaptation coefficient being less than the second texture adaptation coefficient; The degree of synergy between crop transpiration and nutrient demand coefficient is used as an indicator of the synergy of demand parameters: When the degree of synergy is less than or equal to the first synergy threshold, it is determined that the synergy of the demand parameters is poor, and the texture adaptation coefficient needs to be adjusted. If the soil texture is sandy soil, the second texture adaptation coefficient is used to correct the nutrient demand coefficient; if it is clay soil, the first texture adaptation coefficient is used to correct the nutrient demand coefficient. When the degree of synergy is greater than the first synergy threshold and less than or equal to the second synergy threshold, the synergy of the demand parameter is determined to be moderate, the coefficient after the previous round of adjustment is maintained, and the crop transpiration evaporation is finely adjusted by ±5%. When the degree of synergy is greater than the second synergy threshold, the synergy of the demand parameters is determined to be excellent, and the crop transpiration and evaporation and the nutrient demand coefficient are directly used as the final crop water and fertilizer demand parameters. The revised demand parameters must meet the following conditions: crop transpiration and evaporation must be between the minimum and maximum crop transpiration and evaporation, and nutrient demand coefficient must be between the minimum and maximum nutrient demand coefficient. If the parameters are outside these ranges, they must be recalculated.

6. The method for real-time data acquisition and water and fertilizer regulation using a multi-layer soil sensing and demand-supply model according to claim 5, characterized in that, The generation and encrypted transmission of control instructions include: Based on the final water and fertilizer demand parameters, the types of regulation instructions are classified as: water regulation instructions, nutrient regulation instructions, and comprehensive regulation instructions. Pre-set the first, second, and third command priorities in ascending order; determine the priorities based on the crop growth stage. When the crop is in the grain-filling stage, set the water regulation command priority as the first command priority and the nutrient regulation command priority as the second command priority; When the crop is in the seedling stage, set the nutrient regulation command priority as the first command priority and the water regulation command as the second command priority; When soil parameters exceed the threshold simultaneously, a comprehensive control instruction is generated with the priority of the first instruction. The control instructions are encrypted using a session key generated by a distributed key management module. After encryption, the instructions are transmitted to the water and fertilizer execution terminal via a long-distance radio gateway. The validity period of the instructions is set, and if the instructions are not executed before the validity period expires, they are re-encrypted and sent.

7. The method for real-time data acquisition and water and fertilizer regulation using a multi-layer soil sensing and demand-supply model according to claim 6, characterized in that, The distributed key management provides private key decryption and establishes a mapping relationship between data and time and location, including: Assign a unique public and private key to each monitoring area; After receiving the encrypted data packet, the edge processing terminal calls the corresponding area private key stored locally to decrypt it and extract the timestamp and location coordinates from the data packet; Preset time synchronization threshold and position deviation threshold: When the time difference between the decrypted timestamp and the edge processing system is less than or equal to the time synchronization threshold, the time synchronization is considered valid; otherwise, the time is marked as abnormal, and a request is made to resend the data packet with the timestamp. When the deviation between the decrypted location coordinates and the preset monitoring area coordinates is less than or equal to the location deviation threshold, the location mapping is deemed valid; otherwise, the location is marked as abnormal, and the sensor module positioning calibration is triggered. A three-dimensional mapping table of timestamps, location coordinates, and soil parameters is created for valid data and stored in a distributed database, supporting retrieval by time and location.

8. The method for real-time soil multi-layer sensing and water and fertilizer regulation based on demand-supply ratio model according to claim 7, characterized in that, The abnormal data pre-labeling subunit identifies out-of-range values ​​in the collected data, including: Based on the basic soil information, the conventional fluctuation threshold ranges for each parameter are preset: soil moisture fluctuation threshold range, soil electrical conductivity fluctuation threshold range, soil temperature fluctuation threshold range, and crop stem flow rate fluctuation threshold range. Each of the real-time collected parameter values ​​is evaluated individually: When the parameter value is within the normal fluctuation threshold range of the corresponding parameter, it is marked as normal data and included in the valid dataset; When a parameter value is less than the minimum value of the corresponding parameter fluctuation threshold range or greater than the maximum value of the corresponding parameter fluctuation threshold range, it is marked as out-of-range data and the abnormal type is recorded. The threshold for abnormal frequency is set to three times. If the same parameter is collected out of range three times in a row, the sensor module is judged to be faulty and a fault alarm is sent to the management platform. If there is only a single abnormality, it is marked as data to be verified and a re-collection process with a shortened collection interval is triggered.

9. The method for real-time data acquisition and water and fertilizer regulation using a multi-layer soil sensing and demand-supply model according to claim 8, characterized in that, The step of acquiring dynamically changing data based on the multi-parameter sensing module and storing it in a local database includes: A dual-layer storage architecture consisting of a real-time database and a historical database is adopted. The real-time database is used to store dynamically changing data within 24 hours, and the update frequency is consistent with the collection frequency; the historical database is used to archive data by day and retain a one-year historical record. Dynamically changing data includes parameters before regulation, real-time parameters during regulation, and stable parameters after regulation. When storing the data, the regulation instruction identifier and the execution terminal number are attached. Pre-set data compression threshold: When the rate of change of continuously collected parameters is less than or equal to the data compression threshold, the differential compression algorithm is used for storage, and only key data points with a rate of change greater than the data compression threshold are retained; The stored data is encrypted using a public key, and the key is updated synchronously with the distributed key management module every seven days. Historical data is backed up before each update.

10. The method for real-time data acquisition and water and fertilizer regulation using a multi-layer soil sensing and demand-supply ratio model according to claim 9, characterized in that, The continuous data collection based on the final data collection frequency includes: Based on the final data acquisition frequency, the preset frequency adjustment trigger conditions are as follows: When the daily rainfall in the monitored area is greater than or equal to the threshold of the first day's rainfall, the trigger frequency is increased to 1.5 times the final data collection frequency, and the duration is 24 hours. When the rate of change of crop stem flow rate is less than or equal to the second stem flow rate change rate threshold, the trigger frequency is reduced to 0.5 times the final data acquisition frequency, and the duration is twelve hours. Among them, upward adjustment instructions for daily rainfall greater than or equal to the first daily rainfall threshold take precedence over downward adjustment instructions for stem flow rate change less than or equal to the second stem flow rate change threshold. After a frequency adjustment, a frequency change log is generated, recording the reason for the adjustment, the frequencies before and after the adjustment, and the effective time. If the adjusted frequency exceeds the preset range, the range boundary value will be taken as the final frequency. During continuous data collection, the frequency is checked for reasonableness every six hours. If the triggering condition disappears, the original final data collection frequency is restored.

Images