High-standard farmland water and fertilizer integrated intelligent regulation and control method and system

Abstract

The invention provides a high-standard farmland water and fertilizer integrated intelligent regulation and control method and system, and the method comprises the steps: obtaining terrain point cloud data and crop canopy water stress data of a high-standard farmland, generating an elevation gradient distribution feature set, and generating an elevation model according to the set; analyzing the crop canopy moisture stress data, and calculating a moisture stress index; performing spatial position matching on the terrain difference parameters and the micro-terrain difference parameters to generate a terrain compensation parameter set; establishing a time sequence optimization correlation model in the 5G edge computing node, and optimizing a pulse irrigation time sequence factor and a flow distribution weight by using the time sequence optimization correlation model according to the terrain compensation parameter set and the historical irrigation record so as to generate a multi-target irrigation control instruction; performing high-standard farmland water and fertilizer integrated intelligent regulation and control according to the multi-target irrigation control instruction and the water and fertilizer migration rate parameter corresponding to the microtopography difference parameter; water and fertilizer are accurately distributed according to needs, and the irrigation uniformity, the resource utilization efficiency and the crop yield are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of water and fertilizer regulation, and particularly to an intelligent regulation method and system for integrated water and fertilizer in high-standard farmland. Background Art

[0002] In the construction of high-standard farmland, although the land flatness has reached the specification requirements, micro-topographic differences (such as standardized plots with a slope drop ≤ 0.5%) will still cause uneven movement of water and fertilizer in the field. Due to the local elevation differences caused by terrain undulations, problems such as water accumulation in low-lying areas, insufficient water supply in high places, and fertilizer loss in runoff areas are likely to occur during the irrigation process, resulting in uneven absorption of water and nutrients by crop roots and affecting yield and resource utilization efficiency.

[0003] Currently, for the problem of uneven movement of water and fertilizer caused by micro-topographic differences, the existing mainstream technical solutions generate a three-dimensional elevation model through topographic surveying and mapping, combine the data of soil moisture sensors, preset irrigation thresholds for different topographic regions, and control the drip irrigation system to supply water differentially in different regions through a fixed program, and calibrate the parameters manually at regular intervals. Such existing solutions have some limitations: relying on static parameters, resulting in the lack of dynamic response of irrigation decision-making to the actual water demand of crops, and phenomena such as drought in high places and excessive wetness in low places may occur; parameter update depends on manual calibration or sensor data at fixed intervals, and it is impossible to quickly respond to the subtle changes in local micro-topography caused by factors such as tillage and settlement, resulting in the lag of water and fertilizer distribution strategies behind actual needs. Summary of the Invention

[0004] The present invention provides an intelligent regulation method and system for integrated water and fertilizer in high-standard farmland to solve the problems in the prior art that rely on static parameters, resulting in the lack of dynamic response of irrigation decision-making to the actual water demand of crops; parameter update depends on manual calibration or sensor data at fixed intervals, and it is impossible to quickly respond to the subtle changes in local micro-topography caused by factors such as tillage and settlement, resulting in the lag of water and fertilizer regulation strategies behind actual needs.

[0005] In a first aspect, the present invention provides an intelligent regulation method for integrated water and fertilizer in high-standard farmland, including: Obtaining the topographic point cloud data and crop canopy water stress data of high-standard farmland; According to the topographic point cloud data, generating a set of elevation gradient distribution features, and generating an elevation model based on the set of elevation gradient distribution features, where the elevation model includes irrigation low-lying area identification, runoff area boundary parameters, and micro-topography difference parameters; Analyzing the crop canopy water stress data to calculate the water stress index; Performing spatial position matching between the water stress index and the micro-topography difference parameters to generate a terrain compensation parameter set that integrates crop physiological state and topographic features; Establish a timing optimization association model in the 5G edge computing node. According to the terrain compensation parameter set and historical irrigation records, use the timing optimization association model to optimize the pulse irrigation timing factor and flow distribution weight to generate a multi-objective irrigation control instruction; According to the multi-objective irrigation control instruction and the water and fertilizer migration rate parameter corresponding to the micro-topography difference parameter, perform intelligent regulation of water and fertilizer integration in high-standard farmland.

[0006] Optionally, generate an elevation gradient distribution feature set according to the terrain point cloud data, and generate an elevation model according to the elevation gradient distribution feature set. The elevation model includes an irrigation low-lying area identifier, a runoff area boundary parameter, and a micro-topography difference parameter, including: Based on a preset adjacent point quantity threshold, calculate the elevation difference and horizontal distance between each coordinate point in the terrain point cloud data and adjacent coordinate points to generate an elevation gradient distribution feature set. The gradient distribution feature set includes a slope value, a slope direction angle, and a curvature value; Perform spatial interpolation processing on the slope value to generate a slope distribution surface. Mark the closed area where the slope value in the slope distribution surface is less than or equal to the first preset threshold and the coverage area is greater than the second preset threshold as the irrigation low-lying area identifier; Mark the boundary line segment where the slope difference between each coordinate point and adjacent coordinate points in the slope distribution surface exceeds the third preset threshold as the runoff area boundary; Divide the terrain point cloud data to obtain multiple data grid units. Calculate the standard deviation of the slope values of all coordinate points in each data grid unit, and mark the data grid unit where the standard deviation of the slope values is greater than or equal to the fourth preset threshold as the micro-topography difference parameter; Perform associated mapping on the spatial coordinate range of the irrigation low-lying area identifier, the topological connection relationship of the runoff area boundary, and the coding information of the data grid unit corresponding to the micro-topography difference parameter to generate an elevation model.

[0007] Optionally, analyze the crop canopy water stress data to calculate the water stress index, including: Divide the crop canopy water stress data into multiple detection areas, and extract the first reflectivity value and the second reflectivity value of the detection area; Based on the first reflectivity value and the second reflectivity value, generate a normalized reflectivity feature value corresponding to each detection area; Obtain the growth stage code of the crops in the high-standard farmland, and determine the corresponding determination threshold range according to the preset stress determination threshold mapping table; Compare the normalized reflectivity feature value with the determination threshold range to generate a water status label; Calculate the water stress index corresponding to each detection area based on the preset weight coefficient and linear conversion rule corresponding to the water status label.

[0008] In a second aspect, the present invention provides a high-standard farmland integrated water and fertilizer intelligent regulation system, including: An acquisition module for acquiring the terrain point cloud data and crop canopy water stress data of the high-standard farmland; A generation module for generating an elevation gradient distribution feature set according to the terrain point cloud data, and generating an elevation model according to the elevation gradient distribution feature set, where the elevation model includes an irrigation low-lying area identifier, a runoff area boundary parameter, and a micro-topography difference parameter; An analysis module for analyzing the crop canopy water stress data to calculate the water stress index; A matching module for spatially matching the water stress index with the micro-topography difference parameter to generate a terrain compensation parameter set that fuses the crop physiological state and terrain features; An optimization module for establishing a timing optimization association model in a 5G edge computing node, and using the timing optimization association model to optimize the pulse irrigation timing factor and flow distribution weight according to the terrain compensation parameter set and historical irrigation records to generate a multi-objective irrigation control instruction; A regulation module for performing intelligent regulation of the integrated water and fertilizer in the high-standard farmland according to the multi-objective irrigation control instruction and the water and fertilizer migration rate parameter corresponding to the micro-topography difference parameter.

[0009] In a third aspect, the present invention provides a computing device, including a processor and a memory, where a computer program is stored in the memory, and the processor is configured to run the computer program to execute any one of the methods for intelligent regulation of integrated water and fertilizer in high-standard farmland in the first aspect.

[0010] In a fourth aspect, the present invention provides a computer storage medium, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the methods for intelligent regulation of integrated water and fertilizer in high-standard farmland described in any one of the first aspects are implemented.

[0011] In the present invention, topographic point cloud data and crop canopy water stress data of high-standard farmland are obtained; according to the topographic point cloud data, an elevation gradient distribution feature set is generated, and an elevation model is generated based on the elevation gradient distribution feature set, where the elevation model includes irrigation low-lying area identifiers, runoff area boundary parameters, and micro-topography difference parameters; the crop canopy water stress data is analyzed to calculate a water stress index; the water stress index is spatially matched with the micro-topography difference parameters to generate a topographic compensation parameter set that fuses crop physiological states and topographic features; a time-series optimization association model is established in a 5G edge computing node, and according to the topographic compensation parameter set and historical irrigation records, the time-series optimization association model is used to optimize pulse irrigation time-series factors and flow distribution weights to generate multi-objective irrigation control instructions; according to the multi-objective irrigation control instructions and the water and fertilizer migration rate parameters corresponding to the micro-topography difference parameters, intelligent regulation of water and fertilizer integration in high-standard farmland is carried out. The technical solution provided by the present invention synchronously obtains high-precision topographic features and crop physiological state data, provides a multi-dimensional input basis for water and fertilizer regulation, and solves the problem of spatio-temporal inconsistency caused by the separate acquisition of topographic and crop data in traditional methods; quantifies irrigation low-lying areas, runoff paths, and micro-topography difference parameters based on topographic point cloud data, and constructs an elevation model to solve the problem that traditional topographic surveys cannot capture the micro-topography fluctuations of high-standard farmland; solves the problems of subjectivity and insufficient adaptability of traditional visual interpretation or single-band threshold methods through a standardized water stress index; constructs a coupling decision basis for crop water demand and topographic diversion ability by spatially matching and fusing the water stress index and micro-topography difference parameters, and solves the problem of local water and fertilizer imbalance caused by single topographic or physiological index regulation; uses an edge computing node to fuse real-time topographic compensation parameter sets and historical irrigation rules, and dynamically optimizes pulse irrigation time-series factors and flow distribution weights to solve the problem that traditional fixed-time and fixed-quantity irrigation modes cannot adapt to dynamic changes in micro-topography; combines water and fertilizer migration rate parameters to implement spatio-temporal differential execution of irrigation actions, and solves the compound defects of waste in high-leakage areas and insufficient supply in low-permeability areas caused by traditional uniform irrigation. Further, the crop canopy water stress data is divided into detection areas, and the first reflectance value and the second reflectance value are extracted from it to generate a normalized reflectance feature value; by matching the growth stage coding to determine the threshold range, based on the preset weight coefficient and linear conversion rule corresponding to the water state label generated by the normalized reflectance feature value and the determination threshold range, a water stress index is generated. Among them, through the cooperation of the dynamic threshold mechanism and growth stage coding, the defect that traditional fixed threshold methods cannot adapt to changes in the crop growth period is solved; based on the linear mapping of the normalized reflectance feature value and the water state label, a refined grading determination of water stress is realized, significantly improving the spatial adaptability and physiological response accuracy of water and fertilizer regulation decisions in high-standard farmland.

[0012] These aspects or other aspects of the present invention will be more clearly understood in the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0014] Figure 1 It is a flowchart of an intelligent regulation method for integrated water and fertilizer in high-standard farmland provided by an embodiment of the present invention; Figure 2 It is a schematic structural diagram of an intelligent regulation system for integrated water and fertilizer in high-standard farmland provided by an embodiment of the present invention; Figure 3 It is a schematic structural diagram of a computing device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0015] In order to enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention.

[0016] In some processes described in the specification, claims and above-mentioned drawings of the present invention, a plurality of operations appear in a specific order. However, it should be clearly understood that these operations can be executed not in the order in which they appear in this article or in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes can include more or fewer operations, and these operations can be executed in sequence or in parallel. It should be noted that the descriptions such as "first" and "second" in this article are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, nor do they limit that "first" and "second" are of different types.

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0018] Figure 1 An embodiment of the present invention provides a flowchart of an intelligent regulation method for integrated water and fertilizer in high-standard farmland, as Figure 1 shown, the method includes: Regarding the problem of uneven water and fertilizer migration caused by micro-topographic differences in high-standard farmland, the existing technologies rely on static preset parameters and fixed program control, and cannot dynamically respond to the actual water demand of crops and subtle topographic changes, resulting in lagging irrigation decisions and unbalanced water and fertilizer distribution. The core breakthrough of the present invention is as follows: by fusing high-precision topographic point cloud data and crop canopy water stress data for real-time monitoring, a topographic compensation parameter set is constructed, and combined with the time-series optimization correlation model of 5G edge computing, dynamic adaptive adjustment of irrigation strategies is realized. Specifically, first, the elevation model is used to accurately identify micro-topographic differences (such as low-lying areas and runoff areas), and at the same time, the physiological state of crops is quantified through the water stress index, and the topographic characteristics are spatially matched with the water demand of crops to generate a topographic compensation parameter set; secondly, relying on the low-latency characteristics of 5G edge computing nodes, historical irrigation records and the topographic compensation parameter set are integrated in real time to establish a time-series optimization model, and the pulse irrigation time series and flow distribution weight are optimized with multi-objective constraints, so as to break through the limitations of traditional fixed thresholds. This method solves key problems such as lagging static parameters, failure to capture local topographic changes in time, and disconnection between irrigation strategies and actual crop needs through multi-dimensional coupling of topographic-physiological-historical data, realizes accurate on-demand distribution of water and fertilizer in the micro-topographic difference scenario, and improves resource utilization efficiency and crop yield. Based on this, the present invention provides a method for intelligent regulation and control of water and fertilizer integration in high-standard farmland, such as Figure 1 , including: Step 101: Obtain the topographic point cloud data and crop canopy water stress data of the high-standard farmland.

[0019] In this step, the high-standard farmland refers to the farmland that has reached the standards of flat fields, concentrated and contiguous plots, and perfect facilities through land improvement, and meets the land flatness requirements (slope drop ≤ 0.5%). The topographic point cloud data refers to the dense three-dimensional coordinate point set obtained by lidar scanning, which is used to construct an elevation model and reflect the micro-topographic undulation characteristics. The crop canopy water stress data refers to the crop canopy reflectance data obtained by a multispectral sensor, which includes the reflectance values of preset water-sensitive bands and preset reference bands.

[0020] In the embodiment of the present invention, the high-standard farmland is scanned three-dimensionally by an airborne lidar to generate topographic point cloud data including three-dimensional coordinates (X / Y / Z); at the same time, the reflectance spectrum data of the crop canopy is collected by a drone carrying a multispectral sensor to obtain crop canopy water stress data, which is used to characterize the transpiration intensity and water deficit state of the crops.

[0021] Step 102: Generate an elevation gradient distribution feature set according to the topographic point cloud data, and generate an elevation model according to the elevation gradient distribution feature set. The elevation model includes irrigation low-lying area identification, runoff area boundary parameters, and micro-topographic difference parameters.

[0022] In this step, the elevation gradient distribution feature set refers to the set of slope, aspect, and curvature values extracted from the topographic point cloud data, which quantifies the change characteristics of the surface morphology. The elevation model refers to the three-dimensional surface model expressed in the form of a digital matrix, which includes the identification of irrigation low-lying areas, the boundary of the runoff area, and the micro-topography difference parameters. The identification of irrigation low-lying areas refers to the closed area marked in the elevation model where the slope value ≤ the first preset threshold and the coverage area > the second preset threshold, which is prone to waterlogging or salinization. The runoff area boundary parameter refers to the boundary line segment of the surface runoff path delimited according to the slope difference, which is used to predict the high-risk area of water and fertilizer loss. It refers to the standard deviation of the slope calculated in units of grids, which quantifies the local topographic fluctuation intensity (the standard deviation ≥ 0.15 indicates a significant difference). The micro-topography difference parameter refers to the standard deviation of the slope calculated in units of data grids, which quantifies the local topographic fluctuation intensity (the standard deviation of the slope value ≥ 0.15 indicates a significant difference).

[0023] In the embodiment of the present invention, the elevation difference and horizontal distance between each coordinate point and its adjacent points in the topographic point cloud data are calculated to generate an elevation gradient distribution feature set including slope values, aspect angles, and curvature values; subsequently, the slope values are processed by spatial interpolation to generate a slope distribution surface, and the closed area where the slope value ≤ the first preset threshold and the coverage area > the second preset threshold is marked as an irrigation low-lying area; at the same time, the boundary line segment with a slope difference > the third preset threshold is identified as the runoff area boundary. Then, the topographic point cloud data is divided into data grid units, the standard deviation of the slope values within each unit is calculated, and the units with a slope value standard deviation ≥ the fourth preset threshold are marked as micro-topography difference parameters. Finally, all the above results are integrated to generate an elevation model including topographic features.

[0024] Step 103: Analyze the crop canopy water stress data to calculate the water stress index.

[0025] In this step, the water stress index refers to the 0-1 continuous value generated by the reflectance characteristic value and the dynamic threshold, which reflects the degree of crop water deficit.

[0026] In the embodiment of the present invention, first, the crop canopy water stress data is divided into multiple detection areas, the reflectance values of each area are extracted and the normalized reflectance characteristic values are generated; subsequently, in combination with the current growth stage coding of the crop, the corresponding judgment threshold range is matched from the preset stress judgment threshold table; by comparing the normalized reflectance with the judgment threshold range, a water status label is generated; finally, based on the weight coefficient and linear rule corresponding to the water status label, the water stress index of each area is calculated to realize the quantitative evaluation of the water stress degree.

[0027] Step 104: Perform spatial position matching between the water stress index and the micro-topography difference parameter to generate a topographic compensation parameter set that fuses the physiological state of the crop and the topographic features.

[0028] In this step, the crop physiological state refers to the physiological response characteristics exhibited by the crop due to water stress, which is quantified by the difference in spectral reflectance. The terrain feature refers to the difference in surface water diversion capacity described by the micro-topography difference parameters, which affects the transport rate of water and fertilizer. The terrain compensation parameter set refers to the set of terrain compensation parameters (referring to the quantified value generated by the fusion of the water stress index and the micro-topography difference parameters within a single data grid cell, reflecting the intensity of crop water requirement compensation caused by micro-topography differences in the data grid cell) for all data grid cells covering the high-standard farmland, stored in the form of a spatial distribution matrix, including the coordinates, compensation parameter values, and priority labels of each data grid cell, used to identify the areas that need to be preferentially compensated for water and fertilizer and their regulation intensities.

[0029] In the embodiment of the present invention, the first gridded data of the water stress index (constituted by the water stress index) is spatially position-matched with the second gridded data of the micro-topography difference parameters (constituted by the standard deviation of slope values) to obtain the matched data grid cells; according to the water requirement sensitivity parameters and soil water holding capacity parameters corresponding to the crop type, crop physiological weight coefficients and terrain feature weight coefficients are assigned to each matched data grid cell; combining the crop physiological weight coefficient, terrain feature weight coefficient, water stress index, and micro-topography difference parameter corresponding to each data grid cell, a fusion weight value corresponding to each data grid cell is generated; and it is scaled with a preset compensation reference value to generate the terrain compensation parameter set.

[0030] Step 105: Establish a time series optimization association model in the 5G edge computing node. According to the terrain compensation parameter set and historical irrigation records, use the time series optimization association model to optimize the pulse irrigation time series factor and flow distribution weight to generate a multi-objective irrigation control instruction.

[0031] In this step, the historical irrigation records refer to the historical irrigation duration, fertilizer application amount, and corresponding crop yield data stored in the farmland management system. The time series optimization association model refers to a mathematical model that generates optimization decisions by analyzing the dynamic association between historical irrigation records and the real-time terrain compensation parameter set. The pulse irrigation time series factor refers to the parameter that controls the on-off duration of the irrigation device, which is dynamically adjusted according to the crop water requirement peak and micro-topography differences. The flow distribution weight refers to the proportion of the water and fertilizer flow per unit time in each irrigation area, which is optimized and distributed based on the terrain compensation parameter set. The multi-objective irrigation control instruction refers to a triple instruction set including the pulse time series, flow weight, and execution priority, used to coordinate multi-area irrigation operations.

[0032] In an embodiment of the present invention, historical irrigation records are aligned with a terrain compensation parameter set to determine historical growth stage codes; in combination with the irrigation duration change trend corresponding to the codes and the terrain compensation parameter set, a pulsed irrigation timing factor is calculated, and the flow distribution weights between adjacent data grid cells are determined; through a timing optimization association model constructed by a 5G edge computing node, the pulsed irrigation timing factor and the flow distribution weights are optimized using multi-objective constraints to obtain an optimized value of the pulsed irrigation timing factor and an optimized value of the flow distribution weights, and the two are encapsulated and processed to generate a multi-objective irrigation control instruction.

[0033] Step 106: Perform intelligent regulation of water and fertilizer integration in high-standard farmland according to the multi-objective irrigation control instruction and the water and fertilizer migration rate parameter corresponding to the micro-topography difference parameter.

[0034] In this step, the water and fertilizer migration rate parameter refers to the horizontal diffusion rate of water and fertilizer in the soil, which is calculated based on an empirical formula of micro-topography slope and soil porosity.

[0035] In an embodiment of the present invention, the specific implementation process of performing intelligent regulation of water and fertilizer integration is as follows: According to the optimized value of the pulsed irrigation timing factor and the optimized value of the flow distribution weights in the multi-objective irrigation control instruction, in combination with the water and fertilizer migration rate parameter corresponding to the micro-topography difference parameter, a pressure-compensated drip irrigation device is controlled to perform gradient pulsed irrigation. Among them, the acquisition process of the water and fertilizer migration rate parameter is as follows: Based on the micro-topography difference parameter of each data grid cell in the elevation model, a preset soil type database is queried to match the water and fertilizer horizontal diffusion empirical parameters of the corresponding soil type (such as sandy soil, clay soil), and then the water and fertilizer migration rate parameter of each grid cell is calculated according to the product relationship between the standard deviation of the slope and the empirical parameters of the soil type; finally, the irrigation duration and flow are adjusted according to the water and fertilizer migration rate parameter. Among them, for low-permeability areas (rate parameter ≤ 0.03 m / s), the single-pulse irrigation duration is extended, and for high-permeability areas (rate parameter ≥ 0.05 m / s), the duration is shortened and the flow distribution weight is reduced.

[0036] In an embodiment of the present invention, through an elevation model of micro-topography difference and crop water stress, combined with the real-time optimization ability of 5G edge computing, precise matching regulation of the water and fertilizer migration rate in high-standard farmland is realized; compared with the traditional static threshold method, the present invention can dynamically respond to crop water demand changes and micro-topography fluctuations, improve problems such as waterlogging in low-lying areas and nutrient loss in runoff areas, and enhance the uniformity of water and fertilizer distribution and resource utilization efficiency.

[0037] The present invention provides a specific embodiment. In step 102, according to the terrain point cloud data, an elevation gradient distribution feature set is generated, and an elevation model is generated according to the elevation gradient distribution feature set. The elevation model includes an irrigation low-lying area identifier, a runoff area boundary parameter, and a micro-topography difference parameter, and specifically includes the following steps: Step 201: Based on a preset threshold of the number of adjacent points, calculate the elevation difference and horizontal distance between each coordinate point in the terrain point cloud data and its adjacent coordinate points, and generate an elevation gradient distribution feature set, where the gradient distribution feature set includes slope value, aspect angle, and curvature value; In this step, the preset threshold of the number of adjacent points refers to the lower limit of the number of adjacent coordinate points to be selected when calculating the elevation gradient feature of a single point. For example, 8 points are used to ensure the local consistency of slope value calculation. The elevation difference refers to the vertical height difference between two adjacent coordinate points, which is obtained by subtracting the elevation of the adjacent coordinate point from the elevation of the target coordinate point and reflects the undulation degree of the local terrain. The horizontal distance refers to the straight-line distance between two adjacent coordinate points in the plane coordinate system, which is calculated by the Euclidean distance formula of the plane coordinates. The elevation gradient distribution feature set refers to a feature data set including slope value, aspect angle, and curvature value, which is used to describe the geometric shape change of the terrain surface. The slope value refers to a quantitative index of the surface inclination degree. The aspect angle refers to the azimuth angle of the surface inclination direction, with a range of 0° - 360°. The curvature value refers to a quantitative index of the surface bending degree and reflects the concave and convex characteristics of the terrain.

[0038] In the embodiment of the present invention, taking the current coordinate point as the center, select the nearest adjacent coordinate points (such as 8 points), calculate the elevation difference (the elevation of the adjacent coordinate point minus the elevation of the current coordinate point) and horizontal distance (the Euclidean distance of the plane coordinates) between the current point and each adjacent coordinate point. Subsequently, take the average value of the elevation differences of all adjacent coordinate points as the slope value of the current coordinate point (the average elevation difference divided by the average horizontal distance), calculate the weighted average value of the connection direction angles of the adjacent coordinate points as the aspect angle (the weight is the reciprocal of the horizontal distance), and determine the curvature value based on the absolute value of the second derivative of the elevation difference of three consecutive points. Integrate the slope value, aspect angle, and curvature value of each coordinate point into a gradient distribution feature set.

[0039] Step 202: Perform spatial interpolation processing on the slope value to generate a slope distribution surface, and mark the closed areas in the slope distribution surface where the slope value is less than or equal to the first preset threshold and the covered area is greater than the second preset threshold as irrigation low-lying area identifiers; In this step, the slope distribution surface refers to a continuous two-dimensional slope distribution map generated by an interpolation algorithm, and each pixel point corresponds to a slope value. The first preset threshold refers to the upper limit of the slope value for determining the irrigation low-lying area (for example, 1%), which is set based on the land flatness standard of high-standard farmland. The second preset threshold refers to the minimum area requirement for the irrigation low-lying area (for example, 10 square meters), which is used to exclude scattered small-scale low-lying points. The irrigation low-lying area identifier refers to the closed area marked in the slope distribution surface, where the slope value meets the standard and the area meets the requirements, and is prone to water accumulation.

[0040] In the embodiment of the present invention, first, a linear interpolation method is used to convert the slope values of all points into a continuous slope distribution surface covering the entire high-standard farmland plot. In this surface, continuous regions with slope values less than or equal to the first preset threshold (such as 1%) are screened, and the areas of these regions are further calculated. Only the closed regions with a coverage area greater than the second preset threshold (such as 10 square meters) are retained. Finally, the boundary coordinates and the elevation of the central point of these closed regions are recorded as the irrigation low-lying area identifiers.

[0041] Step 203: Mark the boundary line segments where the slope difference between each coordinate point and its adjacent coordinate point in the slope distribution surface exceeds the third preset threshold as the runoff area boundary; In this step, the mutation amplitude of the slope value refers to the absolute difference in slope values between adjacent coordinate points, which is used to detect potential paths of surface runoff. The third preset threshold refers to the lower limit of slope mutation for determining the runoff area boundary (for example, 5%), which is set based on the hydraulic critical slope. The boundary line segment refers to a broken line formed by connecting continuous mutation points, which reflects the natural path dividing line of surface runoff. The runoff area boundary refers to the set of boundary line segments marked, which is used to predict high-risk areas of water and fertilizer loss.

[0042] In the embodiment of the present invention, each point on the slope distribution surface is traversed, and the absolute value of the slope difference between it and its adjacent coordinate points is calculated. If the slope difference of a certain coordinate point exceeds the third preset threshold (such as 5%), it is marked as a mutation point. Subsequently, the continuous mutation points are connected into broken line segments according to the spatial adjacency relationship, and the discrete line segments with a length less than 1 meter are removed. Finally, the starting and ending coordinates and the slope mutation value of the remaining broken line segments are stored as the runoff area boundary.

[0043] Step 204: Divide the terrain point cloud data to obtain multiple data grid units, calculate the standard deviation of the slope values of all coordinate points within each data grid unit, and mark the data grid units with the standard deviation of the slope values greater than or equal to the fourth preset threshold as micro-topography difference parameters; In this step, the data grid unit refers to a rectangular area obtained by dividing the terrain point cloud data according to a regular size (such as 0.1 m × 0.1 m), which serves as the basic unit for micro-topography analysis. The standard deviation of the slope value refers to the index of the dispersion degree of the slope values of all points within a single data grid unit, which reflects the intensity of local terrain fluctuations. The fourth preset threshold refers to the lower limit of the standard deviation for determining significant micro-topography differences (for example, 0.15), which is set based on the micro-topography control requirements of high-standard farmland. The micro-topography difference parameter refers to the encoding and quantization value of the data grid unit with a qualified standard deviation of the slope value, which is used to identify micro-topography regions that require special regulation.

[0044] In the embodiment of the present invention, first, the topographic point cloud data is divided into regular data grid units with a preset size (such as 0.1 m × 0.1 m). Subsequently, the standard deviation of the slope values of all coordinate points within each data grid unit is calculated, and the data grid units with a standard deviation of the slope value greater than or equal to the fourth preset threshold (such as 0.15) are screened out. Finally, the codes (row and column numbers) of these data grid units and the standard deviation values are recorded as micro-topographic difference parameters.

[0045] Step 205: Correlate and map the spatial coordinate range of the identified irrigation low-lying area, the topological connection relationship of the runoff area boundary, and the coding information of the data grid units corresponding to the micro-topographic difference parameters to generate an elevation model.

[0046] In this step, the spatial coordinate range refers to the set of geometric boundary coordinates of the identified irrigation low-lying area, represented by a sequence of polygon vertices. The topological connection relationship refers to the endpoint sharing rule between the boundary line segments of the runoff area, ensuring the connectivity and consistency of the boundary network. The coding information refers to the unique identifier (such as row and column numbers) of the micro-topographic difference parameter grid, used for rapid matching with the spatial position.

[0047] In the embodiment of the present invention, first, the spatial coordinate range of the identified irrigation low-lying area is aligned with the coordinate system of the preset initial model. Subsequently, the topological connection relationship of the runoff area boundary is established to ensure the continuity of the boundary line segments. At the same time, the codes of the data grid units corresponding to the micro-topographic difference parameters (such as G001, G002) are associated with the coordinates of the grid center points. Finally, all the spatial coordinate ranges, topological connection relationships, and coding information are integrated into unified elevation model data, which is input into the preset initial model to generate an elevation model.

[0048] The embodiment of the present invention uses high-precision topographic gradient feature extraction and spatial interpolation technology to accurately identify the irrigation low-lying area and the runoff boundary. Combining with the grid-based micro-topographic difference quantification model, it realizes the dynamic matching and regulation of the water and fertilizer transport rate in high-standard farmland, improves the terrain feature resolution and the recognition accuracy of the micro-topographic difference area, and shortens the irrigation response time.

[0049] The present invention provides a specific embodiment. In step 103, the crop canopy water stress data is analyzed to calculate the water stress index, which specifically includes the following steps: Step 301: Divide the crop canopy water stress data into multiple detection areas, and extract the first reflectance value and the second reflectance value of the detection areas.

[0050] In this step, the detection area refers to an independent analysis unit divided according to a fixed grid size, which is the same as the data grid unit in the elevation model and is used to achieve the spatial alignment of moisture data and terrain data. The first reflectance value refers to the reflectance measurement value of a preset moisture-sensitive band (such as the near-infrared band) extracted from the crop canopy moisture stress data, which reflects the change in the water content of crop tissues. The second reflectance value refers to the reflectance measurement value of a preset reference band (such as the red-edge band) extracted from the crop canopy moisture stress data and is used as a reference value to eliminate environmental interference.

[0051] In an embodiment of the present invention, first, the crop canopy moisture stress data is spatially divided according to a fixed grid size (such as 0.1 m × 0.1 m) that is the same as the data grid unit size in the elevation model, forming a plurality of independent detection areas, and each detection area corresponds to a physical space unit in the high-standard farmland; subsequently, the first reflectance value and the second reflectance value are extracted from the crop canopy moisture stress data of each detection area to ensure that the reflectance data of each detection area strictly corresponds to its spatial position.

[0052] Step 302: Generate a normalized reflectance eigenvalue corresponding to each detection area based on the first reflectance value and the second reflectance value.

[0053] In this step, the normalized reflectance eigenvalue refers to a standardized index generated through difference summation operation, and the calculation formula is (first reflectance value - second reflectance value) / (first reflectance value + second reflectance value), which is used to quantify the crop water status.

[0054] In an embodiment of the present invention, a difference summation operation is performed on the first reflectance value and the second reflectance value of each detection area, that is, the difference is obtained by subtracting the second reflectance value from the first reflectance value, and then the difference is divided by the sum of the first reflectance value and the second reflectance value to generate a normalized reflectance eigenvalue; this operation converts the reflectance difference between the two bands into a standardized eigenvalue, eliminating the influence of the light intensity difference, so that the eigenvalue only reflects the relative change in the crop canopy water status.

[0055] Step 303: Obtain the growth stage code of the crops in the high-standard farmland, and determine the corresponding determination threshold range according to the preset stress determination threshold mapping table.

[0056] In this step, the growth stage coding refers to the stage identifier generated by matching the current date with the preset crop growth cycle, including the sowing date, the number of days in each stage, and the stage coding mapping relationship, which is used to dynamically adjust the moisture determination logic. The preset stress determination threshold mapping table refers to the query table storing the threshold ranges corresponding to different growth stage codings, including fields such as stage coding, minimum threshold, and maximum threshold. The determination threshold range refers to the closed interval numerical range of the normalized reflectance eigenvalue corresponding to the current growth stage coding, which is used to divide the classification boundary of the crop moisture stress state.

[0057] In the embodiment of the present invention, the growth stage coding of the crop is generated by matching the current crop growth days with the preset growth cycle stage division rule; subsequently, the preset stress determination threshold mapping table is queried according to this coding to obtain the corresponding determination threshold range (such as the seedling stage threshold range [0.1, 0.3], the heading stage threshold range [0.2, 0.5]), ensuring that the threshold adapts to the moisture demand characteristics of different growth periods of the crop.

[0058] Step 304: Compare the normalized reflectance eigenvalue with the determination threshold range to generate a moisture status label; In this step, the moisture status label refers to the classification identifier generated according to the comparison result of the normalized reflectance eigenvalue and the determination threshold range, including three discrete labels: severe stress, moderate stress, and normal status.

[0059] In the embodiment of the present invention, the normalized reflectance eigenvalue of each detection area is compared with the determination threshold range. If the normalized reflectance eigenvalue is less than the minimum value of the determination threshold range, it is marked as a severe moisture stress label; if it is within the determination threshold range, it is marked as a moderate moisture stress label; if it is greater than the maximum value of the determination threshold range, it is marked as a normal moisture status label, and the moisture status label is bound to the spatial position of the detection area.

[0060] Step 305: Calculate the moisture stress index corresponding to each detection area based on the preset weight coefficient and linear conversion rule corresponding to the moisture status label.

[0061] In this step, the preset weight coefficient refers to the reference numerical value assigned to each moisture status label, which is used to convert the discrete label into a continuous index value. The linear conversion rule refers to the calculation method of proportionally mapping the eigenvalue to the index value within the threshold interval to achieve a smooth transition from the label to the numerical value.

[0062] This step maps the label to a continuous value between 0 and 1 using a linear conversion rule according to the preset weight coefficients corresponding to the moisture status labels (e.g., severe = 0.9, moderate = 0.6, normal = 0.1). Specifically, the exponential value is calculated by proportional interpolation within the threshold interval corresponding to the moisture status label (for example, the eigenvalue 0.35 within the moderate label interval [0.2, 0.5] is mapped to the exponential value 0.6+(0.35 - 0.2) / (0.5 - 0.2)×(0.9 - 0.6)=0.75), generating the moisture stress index for each detection area.

[0063] For example, for a high-standard farmland (winter wheat planting area, grid size 0.1 m × 0.1 m), first divide the crop canopy moisture stress data into 1500 detection areas, extract the near-infrared (850 nm) and red-edge (720 nm) reflectance values of each area, and calculate the normalized reflectance eigenvalue = (850 - 720) / (850 + 720)=0.15; the current date is the 55th day after sowing, so the growth stage code 02 (jointing stage) is matched, and the corresponding determination threshold range [0.2, 0.5] is queried according to the preset stress determination threshold mapping table; since the normalized reflectance eigenvalue 0.15 < 0.2, it is marked as a severe stress label; according to the preset weight coefficient 0.9 and the linear conversion rule, the corresponding moisture stress index is calculated to be 0.85.

[0064] The embodiment of the present invention solves the misjudgment problem caused by the change of crop growth period in the traditional method; through the collaborative conversion of the normalized reflectance eigenvalue and the moisture status label, taking into account the discrete state recognition and continuous regulation requirements, the spatial distribution of the moisture stress index is accurately coupled with the micro-topography difference parameters of the high-standard farmland, providing a reliable basis for the integrated water and fertilizer regulation.

[0065] The present invention provides a specific embodiment, step 303, obtaining the growth stage code of the crops in the high-standard farmland, and determining the corresponding determination threshold range according to the preset stress determination threshold mapping table, specifically including the following steps: Step 311: Retrieve the pre-stored sowing date data from the high-standard farmland management system, and calculate the crop growth days according to the sowing date data and the actual collection date, where the actual collection date is the system record date when obtaining the crop canopy moisture stress data.

[0066] In this step, the high-standard farmland management system refers to a computer system used to store and manage the basic data of high-standard farmland, including data such as sowing dates, soil properties, and historical irrigation records, and provides data query services through a standardized interface. The pre-stored sowing date data refers to the crop sowing operation date data pre-entered in the high-standard farmland management system, in the format of year-month-day, and is used to calculate the crop growth cycle. The actual acquisition date refers to the specific date when the crop canopy water stress data is obtained by a drone equipped with a multispectral sensor, which is automatically recorded and stored by the data acquisition device. The crop growth days refer to the cumulative number of days from the sowing date to the actual acquisition date, calculated by subtracting the sowing date from the actual acquisition date. The system record date refers to the date stamp automatically generated by the data acquisition device when performing the operation of obtaining the crop canopy water stress data, and is bound and stored with the data file metadata.

[0067] In the embodiment of the present invention, by accessing the database interface of the high-standard farmland management system, the pre-stored sowing date data is read; the system record date automatically recorded when collecting the crop canopy water stress data is obtained as the actual acquisition date; the actual acquisition date and the sowing date data are calculated for the date difference to obtain the crop growth days.

[0068] Step 312: Match the crop growth days with the preset growth cycle stage division rule to extract the target associated stage code corresponding to the crop growth days as the growth stage code of the crop. The preset growth cycle stage division rule includes multiple time intervals, and each time interval includes a start day, an end day, and an associated stage code.

[0069] In this step, the preset growth cycle stage division rule refers to the standard rule for dividing the entire growth period of the crop into multiple consecutive time intervals by days. The target associated stage code refers to the unique identifier determined by matching the crop growth days with the preset growth cycle stage division rule, reflecting the current physiological stage of the crop. The time interval refers to the description of a single growth stage in the preset growth cycle stage division rule, including the start day (the start day of the interval), the end day (the end day of the interval), and the associated stage code. The start day refers to the start day of the time interval of a certain growth stage, starting from the sowing day as the first day for accumulation. The end day refers to the end day of the time interval of a certain growth stage, starting from the sowing day as the first day for accumulation. The associated stage code refers to the unique identifier preset and bound to the time interval, such as seedling stage = 01, jointing stage = 02, which is used to associate the stress determination threshold.

[0070] In an embodiment of the present invention, the number of days of crop growth is compared with the start day and end day of each time interval in sequence. When the number of days of crop growth is greater than or equal to the start day of a certain time interval and less than or equal to its end day, the target associated stage code corresponding to this time interval is extracted as the growth stage code of the crop, and the target associated stage code uniquely identifies the growth stage of the current crop.

[0071] Step 313: Extract the target threshold lower limit number and target threshold upper limit number corresponding to the growth stage code from a preset stress determination threshold mapping table. The stress determination threshold mapping table includes a stage code field, a threshold lower limit number field, and a threshold upper limit number field. The threshold lower limit number field and the threshold upper limit number field respectively represent the determination boundaries of the normalized reflectance eigenvalue corresponding to different growth stages.

[0072] In this step, the target threshold lower limit number refers to the lower limit of the determination boundary of the normalized reflectance eigenvalue extracted from the preset stress determination threshold mapping table and corresponding to the target associated stage code. The target threshold upper limit number refers to the upper limit of the determination boundary of the normalized reflectance eigenvalue extracted from the preset stress determination threshold mapping table and corresponding to the target associated stage code. The stage code field refers to the data column in the preset stress determination threshold mapping table that stores the associated stage codes that are exactly the same as those in the growth cycle stage division rules. The threshold lower limit number field refers to the data column in the preset stress determination threshold mapping table that stores the determination lower limits of the normalized reflectance eigenvalues corresponding to each growth stage. The threshold upper limit number field refers to the data column in the preset stress determination threshold mapping table that stores the determination upper limits of the normalized reflectance eigenvalues corresponding to each growth stage. The determination boundary refers to the numerical interval range of the normalized reflectance eigenvalue, which is used to divide the crop water stress level and is jointly defined by the threshold lower limit number and the threshold upper limit number.

[0073] In an embodiment of the present invention, the preset stress determination threshold mapping table is a structured data table, including a stage code field, a threshold lower limit number field, and a threshold upper limit number field, where the stage code field is exactly the same as the associated stage code; using the target associated stage code as the query key value, retrieve the matching record in this mapping table, and extract the target threshold lower limit number and target threshold upper limit number from the threshold lower limit number field and the threshold upper limit number field of this record respectively.

[0074] Step 314: Combine the target threshold lower limit number and the target threshold upper limit number to generate a determination threshold range corresponding to the growth stage code.

[0075] In an embodiment of the present invention, the extracted lower limit number of the target threshold and the upper limit number of the target threshold are sorted according to the numerical size. If the lower limit number of the target threshold is less than the upper limit number of the target threshold, they are directly combined into a determination threshold range in the form of a closed interval; if the lower limit number of the target threshold is greater than the upper limit number of the target threshold, the values of the two are exchanged and then combined into a closed interval to ensure that the lower limit value of the determination threshold range is always not greater than the upper limit value.

[0076] In the embodiment of the present invention, through the dynamic matching of the crop growth days and the preset growth cycle stage division rules, the adaptive adjustment of the water stress determination threshold is realized, solving the defect that the traditional fixed threshold method cannot adapt to the changes in the crop growth period; the growth stage coding is accurately associated with the determination boundary to ensure the scientificity and reliability of the water status classification; the closed interval combination mechanism of the determination threshold range avoids the logical abnormality caused by the wrong threshold order, providing an accurate decision-making basis for the integrated water and fertilizer regulation of high-standard farmland.

[0077] The present invention provides a specific embodiment. Step 104: Perform spatial position matching on the water stress index and the micro-topography difference parameter to generate a topographic compensation parameter set that fuses the crop physiological state and topographic features, which specifically includes the following steps: Step 401: Perform spatial position matching on the first spatial distribution data of the water stress index and the second spatial distribution data of the micro-topography difference parameter to obtain the matched data grid units.

[0078] In this step, the first spatial distribution data refers to the spatial set of the water stress index calculated by dividing the detection area through step 301. Each detection area corresponds to an index value, and the two-dimensional matrix data arranged according to the spatial coordinates is used to reflect the field-level distribution characteristics of the crop water status. The second spatial distribution data refers to the spatial set of the slope standard deviation values calculated by dividing the data grid units through step 204. Each data grid unit corresponds to a standard deviation value, and the two-dimensional matrix data arranged according to the spatial coordinates is used to quantify the spatial distribution characteristics of the small topographic undulations in high-standard farmland. The matched data grid units refer to the data grid units unified through the spatial position alignment operation, and at the same time include the water stress index and the micro-topography difference parameter values, with the size unchanged.

[0079] In an embodiment of the present invention, first, the first spatial distribution data formed by arranging the water stress index according to spatial coordinates and the second spatial distribution data formed by arranging the slope standard deviation value according to spatial coordinates are obtained; the geographic coordinate systems (such as the WGS84 coordinate system or the local farmland plane coordinate system) of the two spatial distribution data are unified to ensure that the physical positions corresponding to the same coordinate points are consistent; if the size of the detection area is the same as that of the data grid unit, they are directly corresponding one by one according to the coordinates; if the sizes are different, the water stress index is redistributed to the center point of the data grid unit by interpolation method. Specifically: taking the center point of the data grid unit as a reference, search for the water stress indices of all detection areas within a preset radius (such as 0.05 meters) around it; take the arithmetic mean of the searched water stress indices as the matching value of this data grid unit, and associate each matched data grid unit with its corresponding water stress index and microtopography difference parameter to form a matched data grid unit.

[0080] Step 402: Based on the water demand sensitivity parameter and the soil water holding capacity parameter corresponding to the crop type of the high-standard farmland, assign a crop physiological weight coefficient and a topographic feature weight coefficient to each matched data grid unit.

[0081] In this step, the water demand sensitivity parameter refers to a preset value (0 - 1) according to the crop type, which reflects the tolerance of the crop to water shortage. For example, the water demand sensitivity parameter of rice is 0.9, and that of corn is 0.7. The soil water holding capacity parameter refers to a preset value (0 - 1) according to the soil type, which reflects the water retention performance of the soil. For example, the water holding capacity parameter of clay is 0.8, and that of sandy soil is 0.3. The crop physiological weight coefficient refers to the weight value (0 - 1) assigned to the water stress index, which is positively correlated with the water demand sensitivity parameter and is used to amplify the crop water demand signal. The topographic feature weight coefficient refers to the weight value (0 - 1) assigned to the microtopography difference parameter, which is negatively correlated with the soil water holding capacity parameter and is used to strengthen the influence of topographic diversion.

[0082] In an embodiment of the present invention, the preset water demand sensitivity parameter and soil water holding capacity parameter are called according to the crop type planted in the high-standard farmland. The higher the water demand sensitivity parameter, the greater the crop physiological weight coefficient, and the lower the soil water holding capacity parameter, the greater the topographic feature weight coefficient. And through normalization processing, it is ensured that the sum of the two weight coefficients is always 1. For example, the water demand sensitivity parameter of wheat is 0.8, corresponding to the crop physiological weight coefficient of 0.7, and the water holding capacity parameter of sandy soil is 0.3, corresponding to the topographic feature weight coefficient of 0.6.

[0083] Step 403: Combine the crop physiological weight coefficient, topographic feature weight coefficient, water stress index, and microtopography difference parameter corresponding to each matched data grid unit to generate a fusion weight value corresponding to each data grid unit.

[0084] In this step, the fusion weight value refers to the comprehensive parameter value generated by weighted superposition, which reflects both the water requirement urgency of the crop and the efficiency of water and fertilizer transport in the terrain.

[0085] In the embodiment of the present invention, the water stress index of each data grid unit is multiplied by the crop physiological weight coefficient, and the micro-topography difference parameter is multiplied by the terrain feature weight coefficient, and the product results of the two are added to generate the fusion weight value. The calculation formula is the product of the water stress index and the crop physiological weight coefficient plus the product of the micro-topography difference parameter and the terrain feature weight coefficient. If the calculation result exceeds the range of 0 to 2, it is truncated according to the boundary value.

[0086] Step 404: Perform proportional scaling processing on the fusion weight value based on a preset compensation reference value to generate a terrain compensation parameter set.

[0087] In this step, the preset compensation reference value refers to the reference numerical value set based on historical irrigation effects, which is used to scale the fusion weight value to the actual control dimension (such as the multiple of irrigation duration).

[0088] In the embodiment of the present invention, the preset compensation reference value is extracted from the historical irrigation records as the control dimension reference, and the fusion weight value is divided by this reference value to complete the proportional scaling. For example, when the preset compensation reference value is 0.5, the terrain compensation parameter corresponding to the fusion weight value of 0.8 is 1.6. Finally, the scaled terrain compensation parameters corresponding to each data grid unit are associated with the spatial coordinates to generate a terrain compensation parameter set covering the entire field block.

[0089] The embodiment of the present invention solves the problem of regulation deviation caused by the isolated analysis of (water stress index) and terrain features (micro-topography difference parameters) in the traditional method, and realizes the precise matching of the water and fertilizer requirements of high-standard farmland and the terrain diversion ability.

[0090] The present invention provides a specific embodiment. In step 105, a time-series optimization association model is established in the 5G edge computing node. According to the terrain compensation parameter set and historical irrigation records, the time-series optimization association model is used to optimize the pulse irrigation time-series factor and flow distribution weight to generate a multi-objective irrigation control instruction, which specifically includes the following steps: Step 501: Align the terrain compensation parameter set with the time-series data of the historical irrigation records to obtain the aligned historical irrigation records.

[0091] In this step, the historical irrigation records refer to the irrigation operation data stored in the past, including fields such as historical irrigation dates, single irrigation duration, and regional flow distribution ratios, which are used to analyze historical rules. The time-series data refers to the set of historical irrigation records arranged in chronological order, and its time stamps are strictly associated with the crop growth stages.

[0092] In an embodiment of the present invention, according to the terrain compensation parameter set, the terrain compensation parameters of each data grid unit and the corresponding spatial coordinates of the unit are read in the order of data grid unit coding. From the irrigation log database of the high-standard farmland management system, complete historical irrigation records including historical irrigation dates, single irrigation durations, regional division codes, and flow distribution ratio fields in the most recent three years are screened in reverse chronological order to generate time series data. The historical irrigation dates are aligned with the crop growth days corresponding to the terrain compensation parameters through time axis matching. The spatial coordinates of each data grid unit in the terrain compensation parameter dataset are converted into a coordinate system (such as WGS84) consistent with the regional division code, and the regional division code is mapped to the current data grid unit coding system to ensure that each historical irrigation record strictly corresponds to the current data grid unit in the spatio-temporal dimension, and finally the aligned historical irrigation records are generated.

[0093] Step 502: Based on the aligned historical irrigation records and the preset growth cycle stages, generate historical growth stage codes corresponding to the historical irrigation dates.

[0094] In this step, the aligned historical irrigation records refer to the subset of historical data associated with the data grid unit corresponding to the current terrain compensation parameters through spatio-temporal alignment operations to ensure spatio-temporal dimension consistency. The preset growth cycle stages refer to the predefined rules for dividing the crop growth period, including sowing dates, the number of days in each stage, and the mapping relationship with stage codes (such as 01 - seedling stage, 02 - jointing stage). The historical irrigation dates refer to the dates when irrigation operations were actually performed in the past, which are used to match the historical growth stages of the crops. The historical growth stage codes refer to the stage identifiers calculated from the historical irrigation dates and sowing dates, reflecting the growth period in which the historical irrigation operations occurred.

[0095] In an embodiment of the present invention, based on the historical irrigation dates and the preset growth cycle stages in the aligned historical irrigation records, the number of days between the historical irrigation dates and the historical sowing dates is calculated, and the number of days is matched with the time intervals of the preset growth cycle stages (for example, an interval of 70 days corresponds to the heading stage code 03) to generate historical growth stage codes corresponding to the historical irrigation dates, which are used to identify the crop growth stages in which the historical irrigation operations occurred.

[0096] Step 503: According to the change trend of irrigation duration corresponding to the historical growth stage codes, combined with the terrain compensation parameter set, calculate the pulsed irrigation timing factor.

[0097] In this step, the change trend of irrigation duration refers to the historical statistical law of single irrigation duration within a specific growth stage (such as the duration of the heading stage increasing by 5% week by week). The pulsed irrigation timing factor refers to the parameter that controls the opening duration of a single irrigation, and the initial value is calculated based on the historical duration and the current terrain compensation parameters.

[0098] In an embodiment of the present invention, according to the historical growth stage code, the variation trend of the irrigation duration corresponding to the growth stage is queried (for example, the historical average single irrigation duration during the heading stage is 120 seconds). Combining with the average value of the terrain compensation parameters in the current terrain compensation parameter set (for example, the average value of the terrain compensation parameters is 1.3), the historical average irrigation duration is multiplied by the average value of the terrain compensation parameters to generate an initial pulsed irrigation timing factor (for example, 120 seconds × 1.3 = 156 seconds), which is used as the benchmark control parameter for pulsed irrigation.

[0099] Step 504: Generate a spatial variation gradient according to the parameter difference rate between adjacent data grid cells in the terrain compensation parameter set, so as to assign a flow distribution weight to each matched data grid cell.

[0100] In this step, the parameter difference rate is the change rate of the terrain compensation parameters between adjacent data grid cells in the terrain compensation parameter set, which is used to quantify the spatial difference degree of the terrain diversion ability and reflect the significant terrain undulation in this area. The spatial variation gradient refers to the change rate of the terrain compensation parameters between adjacent data grid cells, which reflects the local terrain diversion difference. The flow distribution weight refers to the proportion of the water and fertilizer flow assigned to each data grid cell, which is positively correlated with the spatial variation gradient and is used to compensate for the terrain diversion difference.

[0101] In an embodiment of the present invention, each data grid cell in the terrain compensation parameter set is traversed, and the parameter difference rate between it and the adjacent data grid cells is calculated (the difference between the terrain compensation parameters between adjacent data grid cells divided by the unit horizontal spacing). All the parameter difference rates are normalized to the range of 0 - 1 to obtain the normalized gradient value. The larger the gradient value, the more significant the change in the terrain diversion ability. According to the positive correlation relationship, a flow distribution weight positively correlated with the normalized gradient value is assigned to each data grid cell (for example, the normalized gradient value of 0.8 corresponds to the flow distribution weight of 0.75). More water and fertilizer flow is preferentially allocated to the high-weight area to compensate for the runoff loss caused by the terrain.

[0102] Step 505: Construct a timing optimization association model in the 5G edge computing node. According to the terrain compensation parameter set and the historical irrigation records, using the timing optimization association model, multi-objective constraint optimization is performed on the pulsed irrigation timing factor and the flow distribution weight to generate an optimized value of the pulsed irrigation timing factor and an optimized value of the flow distribution weight.

[0103] In this step, the optimized value of the pulsed irrigation timing factor refers to the final single irrigation duration parameter generated through multi-objective optimization, which satisfies terrain adaptation and resource constraints. The optimized value of the flow distribution weight refers to the final regional flow ratio parameter generated through multi-objective optimization, which balances terrain diversion and global uniformity.

[0104] In the embodiment of the present invention, a terrain compensation parameter set (reflecting the global water demand situation) and historical irrigation records (such as the average duration of a single irrigation corresponding to the growth stage code) are loaded into the 5G edge computing node, and multi-objective constraint conditions such as setting the upper limit of the daily irrigation total amount are set. By iteratively optimizing the pulse irrigation timing factor and the flow distribution weight, an optimized value of the pulse irrigation timing factor and an optimized value of the flow distribution weight are generated to simultaneously meet the terrain diversion requirements, historical law adaptability, and total resource limitations.

[0105] Step 506: Perform instruction encapsulation processing on the optimized value of the pulse irrigation timing factor and the optimized value of the flow distribution weight to generate a multi-objective irrigation control instruction.

[0106] In the embodiment of the present invention, the optimized value of the pulse irrigation timing factor (such as 138 seconds) and the optimized value of the flow distribution weight (such as 0.48) are encapsulated according to the data grid unit code (such as G12, B07) to generate a machine-readable multi-objective irrigation control instruction, which includes the code, timing parameters, and flow parameters of each data grid unit, driving the irrigation system to perform differential regulation. This instruction is sent to the irrigation control system for execution through the 5G network.

[0107] The embodiment of the present invention realizes the spatio-temporal collaborative decision-making of water and fertilizer regulation in high-standard farmland by aligning and fusing historical irrigation rules and terrain compensation parameters in space and time, combined with the real-time optimization ability of 5G edge computing. Compared with traditional methods, the irrigation duration and flow distribution can simultaneously adapt to the water demand changes during the crop growth period and the micro-topography diversion differences, improving the water and fertilizer utilization rate and the irrigation uniformity in areas with complex topography.

[0108] The present invention provides a specific embodiment. In step 505, a timing optimization association model is constructed in the 5G edge computing node. According to the terrain compensation parameter set and historical irrigation records, using the timing optimization association model, multi-objective constraint optimization is performed on the pulse irrigation timing factor and the flow distribution weight to generate an optimized value of the pulse irrigation timing factor and an optimized value of the flow distribution weight, which specifically includes the following steps: Step 511: Based on historical irrigation records and growth stage codes, statistically calculate the average duration of a single irrigation corresponding to each growth stage code and the fluctuation range of the flow distribution ratio.

[0109] In this step, it refers to the arithmetic mean of the single irrigation durations in all historical irrigation records under the same growth stage code, reflecting the standard irrigation duration of this stage. The fluctuation range of the flow distribution ratio refers to the interval formed by the historical maximum and minimum values of the flow distribution ratio under the same growth stage code, which is used to limit the reasonable range of weight adjustment.

[0110] In the embodiments of the present invention, first, group the historical irrigation records according to the growth stage codes (such as the seedling stage code 01 and the heading stage code 03), calculate the arithmetic mean of all single irrigation durations within each group as the mean single irrigation duration of this stage, and at the same time, count the maximum and minimum values of the flow distribution ratio within each group to generate the fluctuation range of the flow distribution ratio (such as the flow ratio fluctuation range at the heading stage is 0.5 - 0.7).

[0111] Step 512: Set multi-objective constraint conditions, where the multi-objective constraint conditions include the upper limit of the daily irrigation total, the flow difference threshold between adjacent crop planting areas, and the equipment operation energy consumption threshold.

[0112] In this step, the multi-objective constraint conditions refer to the composite limit conditions that need to be satisfied simultaneously during the optimization process, including the total amount of resources, regional balance, and equipment capacity. The upper limit of the daily irrigation total refers to the maximum allowable amount of water and fertilizer irrigation per day, which is calculated based on the rated water supply of the water pump and the farmland area. The flow difference threshold refers to the maximum allowable difference in the flow distribution ratio per unit time between adjacent crop planting areas, which is used to prevent local over-irrigation or under-irrigation. The equipment operation energy consumption threshold refers to the maximum allowable operation duration or power consumption of the irrigation equipment per day, which is set based on the equipment performance parameters.

[0113] In the embodiments of the present invention, set the upper limit of the daily irrigation total (such as 500 tons) according to the maximum daily water supply of the water pump in the high-standard farmland, set the flow difference threshold (such as the flow ratio difference between adjacent areas does not exceed 10%) based on the difference in the water demand characteristics of crops in adjacent farmland plots, and set the equipment operation energy consumption threshold (such as the maximum operation duration per day is 8 hours) based on the rated power of the irrigation equipment and the operation time limit.

[0114] Step 513: Based on the mean single irrigation duration and the fluctuation range of the flow distribution ratio, calculate the global adaptation deviation value of the pulse irrigation timing factor and the flow distribution weight.

[0115] In this step, the global adaptation deviation value refers to the deviation degree index of the current parameter from the historical law, which is used to quantify the deviation degree of the current parameter from the historical law.

[0116] In the embodiments of the present invention, add the absolute value of the difference between the current pulse irrigation timing factor (such as 160 seconds) and the mean single irrigation duration of the corresponding growth stage (such as 130 seconds), and the absolute value of the difference between the current flow distribution weight (such as 0.7) and the upper limit of the historical flow distribution ratio fluctuation range (such as 0.6) to generate the comprehensive deviation value (i.e., |160 - 130| + |0.7 - 0.6| = 30.1).

[0117] Step 514: Based on the global adaptation deviation value and the parameter difference rate, establish a parameter adjustment rule in the timing optimization correlation model. The parameter adjustment rule includes that the adjustment amplitude of the pulse irrigation timing factor is positively correlated with the global adaptation deviation value, and the adjustment direction of the flow distribution weight is positively correlated with the parameter difference rate.

[0118] In this step, the parameter adjustment rule refers to the logical basis for parameter modification in the optimization process.

[0119] In the embodiment of the present invention, define the adjustment amplitude of the pulse irrigation timing factor in the timing optimization correlation model as a preset ratio of the global adaptation deviation value (such as deviation value × 0.1). At the same time, it is stipulated that the adjustment direction of the flow distribution weight needs to be positively correlated with the parameter difference rate of adjacent grid cells in the terrain compensation parameter set (such as a difference rate of 0.8 corresponding to a weight adjustment amplitude of +0.15), and the adjusted weight shall not exceed the historical flow distribution ratio fluctuation range.

[0120] Step 515: According to the parameter adjustment rule, use the timing optimization correlation model to iteratively optimize the pulse irrigation timing factor and the flow distribution weight until the multi-objective constraint conditions are met, and obtain the optimized value of the pulse irrigation timing factor and the optimized value of the flow distribution weight.

[0121] In the embodiment of the present invention, gradually adjust the pulse irrigation timing factor and the flow distribution weight according to the parameter adjustment rule, and detect whether the daily total amount, regional flow difference, and energy consumption constraints are met after each adjustment. If any constraint is exceeded, reduce the adjustment amplitude proportionally (such as the original adjustment amplitude × 0.5) and recalculate until all constraint conditions are met, and obtain the optimized value of the pulse irrigation timing factor and the optimized value of the flow distribution weight.

[0122] The embodiment of the present invention realizes multi-objective dynamic optimization in the 5G edge computing node by integrating terrain compensation parameters and historical irrigation rules, solves the problems of uneven resource allocation and the disconnection from historical experience in traditional methods; the optimized values of the pulse irrigation timing factor and the flow distribution weight simultaneously meet terrain diversion compensation, historical rule adaptation, and resource constraints, improving the water and fertilizer utilization rate and the irrigation uniformity of the entire field.

[0123] Figure 2 The following is a schematic structural diagram of a high-standard farmland water and fertilizer integration intelligent control system provided by the embodiment of the present invention, as Figure 2 shown. The system includes: An acquisition module 21, configured to acquire the terrain point cloud data and crop canopy water stress data of the high-standard farmland; A generation module 22, configured to generate a set of elevation gradient distribution features based on the terrain point cloud data, so as to generate an elevation model according to the set of elevation gradient distribution features, where the elevation model includes an irrigation low-lying area identifier, a runoff area boundary parameter, and a micro-topography difference parameter; An analysis module 23, configured to analyze the crop canopy water stress data to calculate a water stress index; A matching module 24, configured to perform spatial position matching on the water stress index and the micro-topography difference parameter to generate a terrain compensation parameter set that fuses the physiological state of the crop and the terrain features; An optimization module 25, configured to establish a timing optimization association model in a 5G edge computing node, and optimize the pulse irrigation timing factor and the flow distribution weight according to the terrain compensation parameter set and the historical irrigation records by using the timing optimization association model to generate a multi-objective irrigation control instruction; A regulation module 26, configured to perform intelligent regulation of the integration of water and fertilizer in high-standard farmland according to the multi-objective irrigation control instruction and the water and fertilizer migration rate parameter corresponding to the micro-topography difference parameter.

[0124] Figure 2 The intelligent regulation system for the integration of water and fertilizer in high-standard farmland described above can execute Figure 1 The intelligent regulation method for the integration of water and fertilizer in high-standard farmland described in the embodiments shown, and its implementation principle and technical effects will not be elaborated. For the intelligent regulation system for the integration of water and fertilizer in high-standard farmland in the above embodiments, the specific manners in which each module and unit perform operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0125] In a possible design, Figure 2 The intelligent regulation system for the integration of water and fertilizer in high-standard farmland described in the embodiments shown can be implemented as a computing device, such as Figure 3 shown, and the computing device may include a storage component 31 and a processing component 32; The storage component 31 stores one or more computer instructions, where the one or more computer instructions are called and executed by the processing component 32.

[0126] The processing component 32 is used for: acquiring the terrain point cloud data and crop canopy water stress data of high-standard farmland; generating an elevation gradient distribution feature set according to the terrain point cloud data, so as to generate an elevation model according to the elevation gradient distribution feature set, where the elevation model includes an irrigation low-lying area identifier, a runoff area boundary parameter, and a micro-topography difference parameter; parsing the crop canopy water stress data to calculate a water stress index; performing spatial position matching between the water stress index and the micro-topography difference parameter to generate a terrain compensation parameter set that fuses the physiological state of the crop and the terrain features; establishing a timing optimization association model in the 5G edge computing node, and optimizing the pulse irrigation timing factor and the flow distribution weight according to the terrain compensation parameter set and the historical irrigation record by using the timing optimization association model to generate a multi-objective irrigation control instruction; performing intelligent regulation of the integration of water and fertilizer in high-standard farmland according to the multi-objective irrigation control instruction and the water and fertilizer migration rate parameter corresponding to the micro-topography difference parameter.

[0127] Among them, the processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for executing the above method.

[0128] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0129] Of course, the computing device may also necessarily include other components, such as input / output interfaces, display components, communication components, etc.

[0130] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above peripheral interface module may be an output device, an input device, etc.

[0131] The communication component is configured to facilitate communication between the computing device and other devices in a wired or wireless manner, etc.

[0132] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server. The above-mentioned processing components, storage components, etc. can be basic server resources leased or purchased from a cloud computing platform.

[0133] An embodiment of the present invention also provides a computer storage medium storing a computer program, which when executed by a computer can implement the above-mentioned Figure 1 intelligent regulation method for integrated water and fertilizer in high-standard farmland shown in the embodiment.

[0134] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0135] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0136] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0137] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments or perform equivalent replacements for some of the technical features. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent regulation method for integrated water and fertilizer in high-standard farmland, characterized in that, Including: Obtaining terrain point cloud data and crop canopy water stress data of high-standard farmland; Generating an elevation gradient distribution feature set according to the terrain point cloud data, and generating an elevation model according to the elevation gradient distribution feature set, where the elevation model includes an irrigation low-lying area identifier, a runoff area boundary parameter, and a micro-topography difference parameter; Analyzing the crop canopy water stress data to calculate a water stress index; Performing spatial position matching between the water stress index and the micro-topography difference parameter to generate a terrain compensation parameter set integrating crop physiological state and terrain features; Establishing a timing optimization association model in a 5G edge computing node, and optimizing the pulse irrigation timing factor and flow distribution weight by using the timing optimization association model according to the terrain compensation parameter set and historical irrigation records to generate a multi-objective irrigation control instruction; Performing intelligent regulation of water and fertilizer integration in high-standard farmland according to the multi-objective irrigation control instruction and the water and fertilizer migration rate parameter corresponding to the micro-topography difference parameter.

2. The method according to claim 1, wherein Generating an elevation gradient distribution feature set according to the terrain point cloud data, and generating an elevation model according to the elevation gradient distribution feature set, where the elevation model includes an irrigation low-lying area identifier, a runoff area boundary parameter, and a micro-topography difference parameter, including: Calculating the elevation difference and horizontal distance between each coordinate point and adjacent coordinate points in the terrain point cloud data based on a preset adjacent point quantity threshold to generate an elevation gradient distribution feature set, where the gradient distribution feature set includes a slope value, a slope direction angle, and a curvature value; Performing spatial interpolation processing on the slope value to generate a slope distribution surface, and marking a closed area in the slope distribution surface where the slope value is less than or equal to a first preset threshold and the coverage area is greater than a second preset threshold as an irrigation low-lying area identifier; Marking a boundary line segment where the slope difference between each coordinate point and adjacent coordinate points in the slope distribution surface exceeds a third preset threshold as a runoff area boundary; Dividing the terrain point cloud data to obtain multiple data grid units, calculating the standard deviation of the slope values of all coordinate points in each data grid unit, and marking the data grid unit where the standard deviation of the slope values is greater than or equal to a fourth preset threshold as a micro-topography difference parameter; Performing association mapping on the spatial coordinate range of the irrigation low-lying area identifier, the topological connection relationship of the runoff area boundary, and the coding information of the data grid unit corresponding to the micro-topography difference parameter to generate an elevation model.

3. The method according to claim 1, wherein Analyzing the crop canopy water stress data to calculate a water stress index, including: Dividing the crop canopy water stress data into multiple detection areas, and extracting a first reflectivity value and a second reflectivity value of the detection areas; Generating a normalized reflectivity feature value corresponding to each detection area based on the first reflectivity value and the second reflectivity value; Obtaining the growth stage code of the crops in the high-standard farmland, and determining a corresponding determination threshold range according to a preset stress determination threshold mapping table; Comparing the normalized reflectivity feature value with the determination threshold range to generate a water status label; Calculate the water stress index corresponding to each detection area based on the preset weight coefficient and linear conversion rule corresponding to the water status label.

4. The method according to claim 3, wherein Obtain the growth stage code of the crops in the high-standard farmland, and determine the corresponding determination threshold range according to the preset stress determination threshold mapping table, including: Retrieve the pre-stored sowing date data from the high-standard farmland management system, and calculate the crop growth days according to the sowing date data and the actual collection date, where the actual collection date is the system record date when obtaining the crop canopy water stress data; Match the crop growth days with the preset growth cycle stage division rule to extract the target associated stage code corresponding to the crop growth days as the growth stage code of the crops. The preset growth cycle stage division rule includes multiple time intervals, and each time interval includes a start day, an end day, and an associated stage code; Extract the target threshold lower limit number and target threshold upper limit number corresponding to the growth stage code from the preset stress determination threshold mapping table. The stress determination threshold mapping table includes a stage code field, a threshold lower limit number field, and a threshold upper limit number field. The threshold lower limit number field and the threshold upper limit number field respectively represent the determination boundaries of the normalized reflectance characteristic values corresponding to different growth stages; Combine the target threshold lower limit number and the target threshold upper limit number to generate a determination threshold range corresponding to the growth stage code.

5. The method according to claim 1, characterized in that Perform spatial position matching on the water stress index and the micro-topography difference parameter to generate a terrain compensation parameter set that fuses the physiological state of the crops and the terrain characteristics, including: Perform spatial position matching on the first spatial distribution data of the water stress index and the second spatial distribution data of the micro-topography difference parameter to obtain the matched data grid unit; Based on the water demand sensitivity parameter and soil water holding capacity parameter corresponding to the crop type in the high-standard farmland, assign a crop physiological weight coefficient and a terrain characteristic weight coefficient to each matched data grid unit; Combine the crop physiological weight coefficient, terrain characteristic weight coefficient, water stress index, and micro-topography difference parameter corresponding to each matched data grid unit to generate a fusion weight value corresponding to each data grid unit; Perform proportional scaling processing on the fusion weight value based on the preset compensation reference value to generate a terrain compensation parameter set.

6. The method according to claim 1, characterized in that Establish a timing optimization association model in the 5G edge computing node. According to the terrain compensation parameter set and the historical irrigation record, use the timing optimization association model to optimize the pulse irrigation timing factor and the flow distribution weight to generate a multi-objective irrigation control instruction, including: Align the time series data of the terrain compensation parameter set with the historical irrigation record to obtain the aligned historical irrigation record; Generate a historical growth stage code corresponding to the historical irrigation date based on the aligned historical irrigation record and the preset growth cycle stage; Calculate the pulse irrigation timing factor according to the irrigation duration change trend corresponding to the historical growth stage code and in combination with the terrain compensation parameter set; Generate a spatial variation gradient based on the parameter difference rate between adjacent data grid cells in the terrain compensation parameter set, so as to assign a flow distribution weight to each matched data grid cell; Construct a timing optimization association model in the 5G edge computing node. According to the terrain compensation parameter set and historical irrigation records, use the timing optimization association model to perform multi-objective constraint optimization on the pulse irrigation timing factor and the flow distribution weight, and generate an optimized value of the pulse irrigation timing factor and an optimized value of the flow distribution weight; Perform instruction encapsulation processing on the optimized value of the pulse irrigation timing factor and the optimized value of the flow distribution weight to generate a multi-objective irrigation control instruction.

7. The method according to claim 6, characterized in that, Construct a timing optimization association model in the 5G edge computing node. According to the terrain compensation parameter set and historical irrigation records, use the timing optimization association model to perform multi-objective constraint optimization on the pulse irrigation timing factor and the flow distribution weight, and generate an optimized value of the pulse irrigation timing factor and an optimized value of the flow distribution weight, including: Based on historical irrigation records and growth stage coding, statistically calculate the average value of the single irrigation duration and the fluctuation range of the flow distribution ratio corresponding to each growth stage coding; Set multi-objective constraint conditions, where the multi-objective constraint conditions include the upper limit of the daily irrigation total, the flow difference threshold between adjacent crop planting areas, and the equipment operation energy consumption threshold; Based on the average value of the single irrigation duration and the fluctuation range of the flow distribution ratio, calculate the global adaptation deviation value of the pulse irrigation timing factor and the flow distribution weight; Based on the global adaptation deviation value and the parameter difference rate, establish a parameter adjustment rule in the timing optimization association model. The parameter adjustment rule includes that the adjustment amplitude of the pulse irrigation timing factor is positively correlated with the global adaptation deviation value, and the adjustment direction of the flow distribution weight is positively correlated with the parameter difference rate; According to the parameter adjustment rule, use the timing optimization association model to iteratively optimize the pulse irrigation timing factor and the flow distribution weight until the multi-objective constraint conditions are met, and obtain the optimized value of the pulse irrigation timing factor and the optimized value of the flow distribution weight.

8. An intelligent control system for integrated water and fertilizer management in high-standard farmland, characterized in that, Including: An acquisition module for acquiring the terrain point cloud data and crop canopy water stress data of high-standard farmland; A generation module for generating an elevation gradient distribution feature set according to the terrain point cloud data, and generating an elevation model according to the elevation gradient distribution feature set. The elevation model includes an irrigation low-lying area identifier, a runoff area boundary parameter, and a micro-topography difference parameter; An analysis module for analyzing the crop canopy water stress data to calculate the water stress index; A matching module for spatially matching the water stress index with the micro-topography difference parameter to generate a terrain compensation parameter set that fuses crop physiological states and terrain features; An optimization module for establishing a timing optimization association model in the 5G edge computing node, and using the timing optimization association model to optimize the pulse irrigation timing factor and the flow distribution weight according to the terrain compensation parameter set and historical irrigation records, so as to generate a multi-objective irrigation control instruction; A regulation module, configured to perform intelligent regulation of integrated water and fertilizer in high-standard farmland according to the multi-objective irrigation control instruction and the water and fertilizer migration rate parameter corresponding to the micro-topography difference parameter.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for intelligent regulation of integrated water and fertilizer in high-standard farmland according to any one of claims 1 to 7.

10. A computer storage medium, characterized in that, A computer program is stored, and when the computer program is executed by a computer, a method for intelligent regulation of integrated water and fertilizer in high-standard farmland according to any one of claims 1 to 7 is implemented.

Images