Intelligent water and fertilizer integrated fertilization device and fertilization parameter optimization method

Abstract

The invention provides an intelligent water and fertilizer integrated fertilization device and a fertilization parameter optimization method. The method comprises the following steps: firstly, constructing a fertilization effect model for describing a relationship between fertilization parameters and crop nutrient absorption effects; on the basis of the fertilization effect model, local sensitivity analysis is carried out on the fertilization parameters according to crop growth stages, the fertilization parameters which have remarkable influence on crop nutrient absorption in each growth stage are identified, and feasible value intervals of the fertilization parameters in each growth stage are limited; constructing a parameter optimization search space according to the feasible value interval of each growth stage, taking a staged fertilization parameter combination as a decision variable, taking a comprehensive index of crop nutrient demand matching degree and fertilizer utilization efficiency as an optimization target, and executing parameter optimization under the constraint of a fertilization effect model to obtain an optimal fertilization parameter combination; and the fertilization control parameters are corrected based on the feedback information in the fertilization execution process. By adopting the scheme of the invention, self-adaptive variable precise fertilization based on the growth stage can be realized, so that the utilization rate of the fertilizer is improved.

Description

Technical Field

[0001] This application relates to the field of intelligent agricultural power machinery technology, and more specifically, to an intelligent integrated water and fertilizer application device and a method for optimizing application parameters. Background Technology

[0002] With the rapid development of intelligent agricultural equipment, precision fertilization technology, represented by intelligent water and fertilizer integration, has become a core component of modern agriculture. Intelligent fertilization devices integrate sensors, controllers, and actuators to achieve programmed control of irrigation and fertilization processes, providing an important hardware foundation for crop growth management.

[0003] However, the optimization capabilities of existing intelligent fertilization technologies still face significant bottlenecks. Current fertilization parameter optimization methods largely rely on static empirical models or global optimization within a single growth cycle, failing to fully consider the dynamic differences in nutrient requirements at different growth stages throughout the crop's entire growth period. Specifically, existing methods typically set parameters such as fertilizer amount and concentration as fixed values ​​or based on the average requirement across the entire cycle, ignoring the fundamental differences in nutrient sensitivity and absorption efficiency at key stages such as seedling, vegetative growth, flowering, and fruiting. This leads to optimization results that often deviate from the actual crop demand curve; insufficient fertilization at one stage may inhibit growth, while excessive fertilization at another stage may cause waste and environmental pollution. Due to the lack of refined identification and constraints on stage-specific parameters, the computational search space of existing optimization models is enormous and inefficient, making it difficult to generate fertilization schemes that truly conform to the crop's physiological rhythm. Therefore, how to achieve precise fertilization based on adaptive variables of growth stages, thereby improving fertilizer utilization, has become a challenge facing the industry. Summary of the Invention

[0004] This application provides an intelligent integrated water and fertilizer application device and a method for optimizing fertilizer application parameters, which can achieve precise fertilization based on adaptive variables according to growth stage, thereby improving fertilizer utilization.

[0005] In a first aspect, this application provides a method for optimizing intelligent fertigation parameters, which is applied to an intelligent fertigation device to optimize crop fertilization parameters. The method includes the following steps: Construct a fertilization effect model to describe the relationship between fertilization parameters and crop nutrient absorption; Based on the fertilization action model, local sensitivity analysis was performed on key fertilization parameters such as fertilization amount, fertilization concentration and fertilization timing at different crop growth stages. The fertilization parameters that significantly affect crop nutrient absorption at each growth stage were identified, and the feasible value range of the fertilization parameters at each growth stage was limited according to the analysis results. A parameter optimization search space is constructed based on the feasible value ranges of each growth stage. The combination of fertilization parameters in stages is used as the decision variable, and the comprehensive index of crop nutrient demand matching degree and fertilizer utilization efficiency is used as the optimization objective. Under the constraints of the fertilization action model, parameter optimization calculation is performed to obtain the optimal combination of fertilization parameters. The corresponding fertilization control parameters are generated based on the optimal fertilization parameter combination, and the fertilization control parameters are dynamically corrected based on the feedback information of crop nutrient status during the fertilization process. Based on the revised fertilization control parameters, liquid fertilizer is applied to the crop root zone simultaneously with irrigation water through a fertilization device, thereby achieving precise fertilization with coordinated water and fertilizer application that matches the crop's growth stage.

[0006] Preferably, constructing a fertilization action model to describe the relationship between fertilization parameters and crop nutrient absorption specifically includes: Based on the migration and transformation process of nutrients in the rhizosphere soil and the absorption mechanism of crop roots, a mathematical model is established to show the relationship between fertilization parameters and changes in rhizosphere soil nutrient concentration. Crop growth stage and environmental factors are used as regulatory variables in the mathematical model. Values ​​are assigned to the kinetic parameters characterizing nutrient absorption in the model to express the dynamic response relationship as the regulatory variables change. By using historical fertilization experiment data, the parameters to be determined in the mathematical model are calibrated and verified, and the verified mathematical model is used as the fertilization effect model.

[0007] Preferably, based on the fertilization model, local sensitivity analysis is performed on key fertilization parameters such as fertilization amount, fertilization concentration, and fertilization timing at different crop growth stages to identify fertilization parameters that significantly affect crop nutrient absorption at each growth stage, specifically including: During each predefined crop growth stage, non-analytical fertilization parameters and environmental conditions are kept constant; The individual fertilization parameters of fertilizer application rate, fertilizer concentration, and fertilization timing are perturbed respectively; Calculate the changes in crop nutrient uptake output by the fertilization model under different perturbation conditions; Based on the magnitude of changes in crop nutrient uptake, the sensitivity of each fertilization parameter was ranked, and the parameters with the highest ranking were identified as key parameters that significantly affect nutrient uptake at the crop's growth stage.

[0008] Preferably, the feasible range of fertilization parameters for each growth stage, based on the analysis results, specifically includes: Sensitivity analysis results of key fertilization parameters at each growth stage were obtained; Based on the results of the sensitivity analysis, the permissible range of fertilization parameter changes during each growth stage was determined; By defining the boundaries of the allowable range, feasible value ranges for fertilization parameters at each growth stage are obtained; The feasible value ranges of fertilization parameters for each growth stage are stored as a set of stage-specific fertilization parameter constraints.

[0009] Preferably, a parameter optimization search space is constructed using feasible value ranges for each growth stage, with the combination of fertilization parameters applied in stages as the decision variable, and a comprehensive index of crop nutrient requirement matching degree and fertilizer utilization efficiency as the optimization objective. Parameter optimization calculations are performed under the constraints of the fertilization action model to obtain the optimal fertilization parameter combination, specifically including: The feasible value ranges of fertilization parameters for each growth stage are obtained and hierarchically organized according to the time sequence of the growth stages to construct a sequentially related stage parameter optimization search space. Within each growth stage, the fertilization amount, concentration, and timing parameters of the growth stage constitute the stage decision variables, and the optimization results of the previous growth stage are used as the parameter constraints of the current growth stage to achieve continuous transfer of parameters between stages. For the current growth stage, under the constraints of the fertilization model, the stage decision variables are combined and solved to construct a stage optimization objective that characterizes the matching degree of crop nutrient demand and fertilizer utilization efficiency at this stage. Parameter optimization calculations are performed stage by stage according to the order of crop growth stages, and the optimal parameter combination obtained after the current stage optimization is completed is used as the fixed boundary condition for the subsequent growth stages. After optimization of all growth stages, the optimal fertilization parameter combinations obtained at each stage are integrated to obtain the optimal fertilization parameter combinations for each stage covering the entire growth period of the crop.

[0010] Preferably, generating the corresponding fertilization control parameters based on the optimal fertilization parameter combination specifically includes: The optimal combination of fertilization parameters is parsed into corresponding fertilization execution parameters; Generate fertilization control instructions based on fertilization execution parameters; The fertilization control commands are configured in a timing sequence to form a fertilization control parameter set; The set of fertilization control parameters is sent to the control unit of the fertilization device.

[0011] Preferably, based on the modified fertilization control parameters, the liquid fertilizer is applied synchronously to the crop root zone via a fertilization device along with irrigation water, specifically including: The dosage and timing of liquid fertilizer application should be controlled according to the revised fertilization control parameters. Liquid fertilizer is delivered simultaneously with irrigation water, and the fertilizer-water mixture is guided to the crop root zone through a fertilization device to complete the fertilization process for the corresponding growth stage.

[0012] Secondly, this application provides an intelligent integrated water and fertilizer application device, comprising: The model building module is used to build a fertilization effect model that describes the relationship between fertilization parameters and crop nutrient absorption. The feature processing module is used to perform local sensitivity analysis on key fertilization parameters such as fertilization amount, fertilization concentration and fertilization timing according to the fertilization action model, and to identify fertilization parameters that have a significant impact on crop nutrient absorption at each growth stage. Based on the analysis results, the module limits the feasible value range of fertilization parameters at each growth stage. The feature processing module is also used to construct a parameter optimization search space with the feasible value range of each growth stage, take the staged fertilization parameter combination as the decision variable, take the comprehensive index of crop nutrient demand matching degree and fertilizer utilization efficiency as the optimization target, and perform parameter optimization calculation under the constraints of the fertilization action model to obtain the optimal fertilization parameter combination. The feature processing module is also used to generate corresponding fertilization control parameters based on the optimal fertilization parameter combination, and to dynamically correct the fertilization control parameters based on feedback information of crop nutrient status during the fertilization process. The fertilization execution module is used to control the application of liquid fertilizer to the crop root zone simultaneously with irrigation water, based on the revised fertilization control parameters, thereby achieving precise fertilization of water and fertilizer in accordance with the crop growth stage.

[0013] Thirdly, this application provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described intelligent water and fertilizer integration fertilization parameter optimization method.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described intelligent water and fertilizer integration fertilization parameter optimization method.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: This application constructs a fertilization mechanism model, combines staged parameter sensitivity analysis and sequential optimization algorithms to generate a global fertilization strategy, and utilizes real-time feedback information for dynamic correction, ultimately achieving adaptive precision fertilization throughout the crop's entire growth period. First, a fertilization action model is constructed to describe the relationship between fertilization parameters and crop nutrient absorption. This model characterizes the response characteristics of the soil-plant system to different fertilization parameters, providing a quantifiable scientific basis and decision boundary for subsequent parameter optimization, ensuring that the optimization process is based on clear mechanistic constraints. Second, based on the fertilization action model, local sensitivity analysis is performed on different growth stages, thereby limiting the feasible value range of fertilization parameters for each stage. By identifying the dominant regulatory factors at each growth stage, the optimization search space is significantly compressed, computational complexity is reduced, and parameter optimization efficiency is improved. Then, a hierarchical search space is constructed based on the staged parameter ranges, and optimization is performed on this basis. The sequential optimization process employs a sequential decision-making strategy, progressively solving the problem from the previous growth stage to the next. Within each stage, under predetermined interval constraints and fertilization model conditions, the optimal fertilization parameters for that stage are determined by using a comprehensive index of nutrient demand matching degree and fertilizer utilization efficiency as the optimization objective. This optimal solution is then passed on as a constraint to the next stage, gradually forming a parameter sequence that precisely matches the dynamic nutrient requirements of the crop, achieving optimal resource allocation over time. Finally, based on feedback information obtained from real-time field monitoring, the fertilization control parameters are dynamically corrected, constructing a closed-loop control mechanism that enables the fertilization strategy to adaptively adjust according to crop growth status and environmental changes. In summary, this application, through the synergistic effect of mechanistic modeling, stage-specific sensitivity analysis, sequential optimization, and feedback control, achieves precise fertilization based on adaptive variables of growth stages, effectively improving fertilizer utilization efficiency and the level of intelligence in fertilization management. Attached Figure Description

[0016] Figure 1 This is an exemplary flowchart of a method for optimizing intelligent water and fertilizer integration fertilization parameters according to some embodiments of this application; Figure 2 This is a schematic flowchart illustrating the construction of a fertilization effect model according to some embodiments of this application; Figure 3 This is a flowchart illustrating the feasible range of fertilization parameters at each growth stage, as shown in some embodiments of this application. Figure 4 This is a structural schematic diagram of an intelligent water and fertilizer integrated fertilization device according to some embodiments of this application; Figure 5 This is a schematic diagram of the structure of a computer device for implementing an intelligent water and fertilizer integration fertilization parameter optimization method according to some embodiments of this application. Detailed Implementation

[0017] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] In existing fertigation technologies, fertilization parameters are typically set based on empirical models or static crop nutrient requirements, failing to fully consider the dynamic changes in nutrient absorption mechanisms, root zone environmental conditions, and the coupling relationship between fertilization parameters and the crop at different growth stages. This results in insufficient matching between fertilization amount, concentration, and timing and the actual nutrient requirements of the crop, leading to technical problems such as low fertilizer utilization efficiency, large fluctuations in nutrient supply, and difficulty in adapting to complex field conditions. To address these issues, a smart fertigation parameter optimization method is proposed, using crop growth stages as the optimization focus. By constructing a fertilization action model to characterize the intrinsic relationship between fertilization parameters and crop nutrient absorption, and combining parameter sensitivity analysis and feasible interval constraints at different growth stages, a staged, hierarchical parameter optimization search space is formed. Under model constraints, the optimal combination of fertilization parameters is obtained, thereby achieving the scientific configuration and dynamic adaptation of fertilization parameters.

[0019] refer to Figure 1 The figure is an exemplary flowchart of an intelligent fertigation parameter optimization method according to some embodiments of this application. The intelligent fertigation parameter optimization method mainly includes the following steps: In step 101, a fertilization effect model is constructed to describe the relationship between fertilization parameters and crop nutrient absorption.

[0020] In some embodiments, reference Figure 2 As shown in the figure, this is a schematic flowchart of constructing a fertilization effect model in some embodiments of this application. In this embodiment, the fertilization effect model used to describe the relationship between fertilization parameters and crop nutrient absorption can be constructed by the following steps: In step 1011, a mathematical model is established based on the migration and transformation process of nutrients in the rhizosphere soil and the crop root absorption mechanism to establish the relationship between fertilization parameters and changes in rhizosphere soil nutrient concentration. In step 1012, crop growth stage and environmental factors are used as regulatory variables in the mathematical model, and values ​​are assigned to the kinetic parameters characterizing nutrient absorption in the model to express the dynamic response relationship as the regulatory variables change. In step 1013, the parameters to be determined in the mathematical model are calibrated and verified using historical fertilization experimental data, and the verified mathematical model is used as the fertilization effect model.

[0021] It should be noted that the fertilization parameters in this application refer to adjustable factors used to control the crop fertilization process, such as fertilizer amount, fertilizer concentration and fertilization sequence; the crop growth stage refers to the continuous period divided according to the significant characteristics of crop morphogenesis and physiological changes, including seedling stage, vegetative growth stage, flowering stage and fruiting and ripening stage; the environmental factors include but are not limited to light, temperature, soil moisture and soil physicochemical properties.

[0022] In practical applications, firstly, based on the migration and transformation process of nutrients in the rhizosphere soil and the crop root absorption mechanism, a mathematical model of the relationship between fertilization parameters and changes in rhizosphere soil nutrient concentration can be established in the following way: A mathematical model based on physicochemical and biological mechanisms is established. This model aims to quantitatively characterize the continuous process from fertilization operations to rhizosphere soil nutrient response and then to crop absorption. The specific modeling process includes: For the migration and transformation process of nutrients in the rhizosphere soil, based on the principle of mass conservation, establishing a partial differential equation with the nutrient concentration in the soil solution as the dependent variable, i.e., a one-dimensional or two-dimensional soil nutrient transport and transformation equation. This equation includes convection terms driven by irrigation water flow, molecular diffusion and mechanical dispersion terms driven by concentration gradients, and source-sink terms describing the adsorption and desorption of nutrients between the solid and liquid phases of the soil, and the nitrification or denitrification reactions occurring in different forms; For the crop root absorption mechanism, a kinetic model is used to describe the rate of nutrient absorption by the roots. This rate is determined by... The nutrient uptake is often expressed as a function of root zone soil nutrient concentration, for example, using the Michaelis-Menten equation. The total nutrient uptake of the crop is obtained by integrating the uptake rate over the entire root zone space defined by the root density distribution function. These two parts are then coupled, with the calculated total root uptake introduced as a negative term in the soil nutrient transport and transformation equation, forming a closed-loop, synchronously solved coupled model. In this coupled model, the fertilization parameters are specified as input or boundary conditions: the fertilization amount determines the total amount of nutrients injected into the soil system; the fertilization concentration and irrigation flow rate together determine the flux intensity of the input nutrients; and the fertilization timing controls the opening, closing, and change of this input flux over time. Finally, the coupled set of partial differential equations reflecting how fertilization parameters affect the dynamics of available root nutrients through soil processes serves as the initial mathematical model describing the relationship between fertilization parameters and changes in root zone soil nutrient concentration.

[0023] Then, crop growth stages and environmental factors are used as control variables in the mathematical model. Values ​​are assigned to the kinetic parameters characterizing nutrient absorption in the model to express the dynamic response relationship as the control variables change. This can be achieved in the following way: First, define the control variables. The crop growth stages are divided into multiple consecutive growth periods according to crop phenology, and a stage coefficient is set for each period (e.g., a growth degree-day index determined based on effective accumulated temperature or calendar days). The environmental factors include at least the root zone soil temperature, light radiation value, and soil volumetric water content acquired in real time by sensors. Next, select the kinetic parameters characterizing root nutrient absorption capacity in the mathematical model as the objects to be dynamically assigned values. These kinetic parameters include at least the maximum absorption rate parameter and the half-saturation constant parameter in the Michaelis-Menten equation describing the relationship between absorption rate and root surface nutrient concentration. Further, based on agricultural... Based on knowledge and historical experimental data, a quantitative relationship function between each kinetic parameter to be assigned and the control variable is established. Specifically, the maximum absorption rate parameter is set as a piecewise linear function of the crop growth stage coefficient, and multiplied by an exponential temperature influence correction coefficient with soil temperature as the variable; the half-saturation constant parameter is set as an inverse proportional function of soil volumetric water content. During model operation, based on the input current crop growth stage information and real-time collected environmental factor data, the specific values ​​of the corresponding kinetic parameters are calculated according to the quantitative relationship function and substituted into the mathematical model. Finally, the mathematical model, which has undergone the above dynamic assignment process and whose absorption kinetic parameters can be adjusted in real time with the crop growth process and environmental conditions, serves as an enhanced mathematical model that can express the dynamic response relationship of nutrient absorption with changes in the control variable.

[0024] Finally, using historical fertilization experimental data, the parameters to be determined in the mathematical model are calibrated and verified. The verified mathematical model can be used as the fertilization effect model in the following way: A pre-constructed historical fertilization experimental dataset is obtained, containing multiple independent experimental records. Each set of records corresponds completely to a combination of fertilization parameters, measured values ​​of root zone soil nutrient concentration monitored over time, and measured values ​​of nutrient uptake at crop maturity. A portion of the data is then selected as a calibration set. The fertilization parameters of each record in the calibration set are input into the enhanced mathematical model, and the model is run to obtain the corresponding predicted sequences of soil nutrient concentration and crop nutrient uptake. A nonlinear least squares optimization algorithm is used, with the optimization objective of minimizing the overall error between the predicted sequence and the measured sequence in the calibration set, to iteratively adjust the model. Several undetermined parameters that cannot be directly determined by the mechanism in the model (such as the longitudinal dispersion coefficient of soil nutrients, nutrient adsorption constant, etc.) are determined until the error converges, thereby obtaining the optimal estimate of the undetermined parameters. Then, model validation is performed, that is, the remaining part of the dataset that was not calibrated is used as the validation set, the fertilization parameters recorded in the validation set are input into the model that has completed parameter calibration, the determination coefficient between the crop nutrient uptake predicted by the model and the actual uptake in the validation set is calculated, and finally the model is accepted: if the calculated determination coefficient is not lower than the preset threshold (e.g., 0.7), the model is judged to have passed the validation, and its prediction accuracy meets the requirements for subsequent optimization and control. Finally, the enhanced mathematical model that has passed the validation and has reliable prediction ability is used as the fertilization action model for subsequent parameter optimization calculation.

[0025] In step 102, based on the fertilization action model, local sensitivity analysis is performed on key fertilization parameters such as fertilization amount, fertilization concentration and fertilization timing at different crop growth stages. The fertilization parameters that have a significant impact on crop nutrient absorption at each growth stage are identified, and the feasible value range of the fertilization parameters at each growth stage is limited according to the analysis results.

[0026] In some embodiments, based on the fertilization model, local sensitivity analysis is performed on key fertilization parameters such as fertilization amount, fertilization concentration, and fertilization timing at different crop growth stages to identify fertilization parameters that significantly affect crop nutrient absorption at each growth stage. These parameters specifically include: During each predefined crop growth stage, non-analytical fertilization parameters and environmental conditions are kept constant; The individual fertilization parameters of fertilizer application rate, fertilizer concentration, and fertilization timing are perturbed respectively; Calculate the changes in crop nutrient uptake output by the fertilization model under different perturbation conditions; Based on the magnitude of changes in crop nutrient uptake, the sensitivity of each fertilization parameter was ranked, and the parameters with the highest ranking were identified as key parameters that significantly affect nutrient uptake at the crop's growth stage.

[0027] It should be noted that parameter perturbation in this application refers to a planned and regular change in the value of a variable in order to quantify the influence of a single variable; local sensitivity analysis refers to a mathematical analysis method that assesses the relative importance of each parameter by perturbing specific input parameters of the system and observing the changes in the model output when other input conditions are constant.

[0028] In practical applications, maintaining non-analytical fertilization parameters and environmental conditions constant within each predefined crop growth stage can be achieved as follows: To conduct accurate local sensitivity analysis, first determine the specific crop growth stage to be analyzed. Within this stage, set other fertilization parameters not involved in the current analysis (e.g., when analyzing fertilization amount, fertilization concentration and timing are non-analytical fertilization parameters) to fixed baseline values. Simultaneously, set the environmental factors driving the fertilization effect model to constant values ​​determined based on typical meteorological conditions or historical average data for this growth stage. This simulates a stable background environment. The purpose of this step is to eliminate interference from fluctuations in non-target variables and ensure that the observed model output changes accurately. The changes are caused by disturbances in a single fertilization parameter being analyzed. Secondly, parameter perturbation for individual fertilization parameters (fertilization amount, concentration, or timing) can be achieved as follows: Under established constant control conditions, select the fertilization parameter to be analyzed (fertilization amount, concentration, or timing) as the perturbation object. Define a perturbation range around its baseline value (e.g., ±10%, ±20%, ±30% of the baseline value). Within this range, select several discrete perturbation level points at equal intervals or according to specific rules. For numerical parameters such as fertilization amount or concentration, the perturbation directly reflects the change in its value. For parameters representing time arrangement, such as timing, the perturbation can be reflected in the fertilization start time. To account for any changes in the timing or duration of fertilization, the fertilization model is run individually at each perturbation level for this parameter, while all other input conditions remain strictly consistent with the constant control conditions. The changes in crop nutrient uptake output by the fertilization model under different perturbation conditions can be calculated as follows: For the currently analyzed fertilization parameter, extract the output of the fertilization model at each perturbation level, i.e., the total amount of nutrients absorbed by the crop throughout the entire growth stage. Use the nutrient uptake output by the model at its baseline value (i.e., the unperturbed constant control conditions) as a reference baseline. Calculate the absolute difference between the nutrient uptake at each perturbation level and the reference baseline. Alternatively, a relative percentage change can be used to obtain a series of quantitative data reflecting how the model output changes with parameter perturbations. Finally, based on the magnitude of changes in crop nutrient uptake, the sensitivity of each fertilization parameter is ranked, and the parameters with the highest ranking are identified as key parameters that significantly affect nutrient uptake at this growth stage of the crop. This can be achieved as follows: After completing the above perturbation and change calculation process for the three parameters of fertilizer amount, fertilizer concentration, and fertilization timing, the nutrient uptake change data sequence corresponding to each parameter is obtained. By comparing these data sequences, for example, by calculating the average or maximum change in nutrient uptake caused by each parameter at all perturbation levels, the three fertilization parameters are quantitatively ranked from largest to smallest according to their degree of influence.The parameters ranking first (or among the top few, such as the first two) in the sorting results are identified as the factors whose numerical changes have the greatest impact on the final nutrient absorption effect during the current crop growth stage.

[0029] In some embodiments, reference Figure 3 As shown in the figure, this is a flowchart illustrating the process of determining the feasible value range of fertilization parameters at each growth stage in some embodiments of this application. In this embodiment, limiting the feasible value range of fertilization parameters at each growth stage based on the analysis results can be achieved by the following steps: In step 1021, the sensitivity analysis results of key fertilization parameters at each growth stage are obtained; In step 1022, based on the results of the sensitivity analysis, the allowable range of fertilization parameter changes in each growth stage is determined; In step 1023, the allowable range is defined by boundary to obtain the feasible value range of fertilization parameters for each growth stage; In step 1024, the feasible value ranges of fertilization parameters for each growth stage are stored as a stage-specific fertilization parameter constraint set.

[0030] It should be noted that the boundary limitation in this application refers to the process of cross-correcting the allowable range of parameter variation with the physical limits of the equipment and the principles of agronomic safety to determine the final usable boundary; the feasible value range is a numerical boundary characteristic index describing the safe and effective optimization of a certain fertilization parameter within a specific crop growth stage.

[0031] In practical application, firstly, based on the sensitivity analysis results, the allowable range of fertilization parameter changes in each growth stage can be determined as follows: For a specific growth stage, its sensitivity analysis results are read, which include the perturbation-response curves and sensitivity rankings of each fertilization parameter; for key parameters with high sensitivity (i.e., ranked high), based on their perturbation-response curves, the parameter change range corresponding to the condition that the model-predicted crop nutrient uptake is not lower than a preset nutrient uptake decline tolerance threshold (e.g., not lower than 90% of the highest predicted value) is selected as the initial allowable range for that parameter; for non-sensitive parameters... For key parameters, a broad empirical range of values ​​is directly given based on conventional agronomic experience (e.g., referring to the recommended range in local planting manuals). Finally, the initial variation range determined for each fertilization parameter within the current growth stage is taken as the allowable range for fertilization parameter variation in that stage. Then, the allowable range is boundary-limited to obtain the feasible value range for fertilization parameters at each growth stage. This can be achieved as follows: Obtain the allowable range of each parameter determined in the previous step, and compare the allowable range of each parameter with the physical execution capability range from the performance parameters of the fertilization device (e.g., the rated flow range of the fertilizer injection pump, the fertilizer mixing system...). The configurable range of mother liquor concentration is mathematically intersected with the agronomical safety upper limit determined based on crop fertilizer tolerance (e.g., the irrigation water limit conductivity to prevent salt damage, the upper limit of single fertilization amount to prevent seedling burn). The range obtained after this double intersection operation is mathematically organized and expressed as a standard closed interval format [minimum, maximum]. This formatted closed interval, constrained by both equipment physical limits and agronomic safety principles, is then used as the feasible value range for the fertilization parameters corresponding to that growth stage. Finally, the feasible value ranges for fertilization parameters at each growth stage are stored as staged fertilization parameters. The fertilizer parameter constraint set can be implemented in the following way: Create a structured dataset, organized according to the predefined growth stage sequence in the entire crop growth period. In this dataset, establish an independent data unit for each growth stage, and within this unit, record the final feasible value range of all fertilizer parameters for that stage using the correspondence of "parameter name - feasible value range". Persist in saving this structured dataset in the form of a configuration file or internal data object. Use the dataset of feasible value ranges of all fertilizer parameters for each stage in the entire growth period as the stage-specific fertilizer parameter constraint set for subsequent global optimization algorithms.

[0032] In step 103, a parameter optimization search space is constructed based on the feasible value ranges of each growth stage. The combination of fertilization parameters in stages is used as the decision variable, and the comprehensive index of crop nutrient demand matching degree and fertilizer utilization efficiency is used as the optimization objective. Under the constraints of the fertilization action model, parameter optimization calculation is performed to obtain the optimal combination of fertilization parameters.

[0033] In some embodiments, a parameter optimization search space is constructed using feasible value ranges for each growth stage, a combination of staged fertilization parameters is used as the decision variable, and a comprehensive index of crop nutrient requirement matching degree and fertilizer utilization efficiency is used as the optimization objective. The optimal fertilization parameter combination is obtained by performing parameter optimization calculations under the constraints of the fertilization model through the following steps: The feasible value ranges of fertilization parameters for each growth stage are obtained and hierarchically organized according to the time sequence of the growth stages to construct a sequentially related stage parameter optimization search space. Within each growth stage, the fertilization amount, concentration, and timing parameters of the growth stage constitute the stage decision variables, and the optimization results of the previous growth stage are used as the parameter constraints of the current growth stage to achieve continuous transfer of parameters between stages. For the current growth stage, under the constraints of the fertilization model, the stage decision variables are combined and solved to construct a stage optimization objective that characterizes the matching degree of crop nutrient demand and fertilizer utilization efficiency at this stage. Parameter optimization calculations are performed stage by stage according to the order of crop growth stages, and the optimal parameter combination obtained after the current stage optimization is completed is used as the fixed boundary condition for the subsequent growth stages. After optimization of all growth stages, the optimal fertilization parameter combinations obtained at each stage are integrated to obtain the optimal fertilization parameter combinations for each stage covering the entire growth period of the crop.

[0034] It should be noted that the parameter optimization search space in this application is a set of structured mathematically feasible domains organized hierarchically according to the growth period sequence, based on the feasible value range of fertilization parameters for each growth stage; the crop nutrient demand matching degree refers to the degree of agreement between the nutrient absorption of the crop stage simulated by the fertilization action model and the theoretical nutrient demand of the crop at that stage; the fertilizer utilization efficiency refers to the ratio between the actual nutrient absorption of the crop through fertilization and the total nutrient input through fertilization at the corresponding growth stage; the optimal fertilization parameter combination is a set of parameters for fertilization amount, concentration, and time sequence at each stage that optimizes the overall growth period.

[0035] In practical applications, firstly, the feasible value ranges of fertilization parameters for each growth stage are obtained, and then organized hierarchically according to the time sequence of the growth stages. The sequentially associated stage-specific parameter optimization search space can be constructed in the following way: From the pre-stored set of stage-specific fertilization parameter constraints, the feasible value ranges of each fertilization parameter in the first growth stage, the second growth stage, and up to the last growth stage are read sequentially. These ranges are organized into a multi-layered data structure according to the order of their respective growth stages. The first layer of data blocks stores the ranges of all parameters in the first growth stage, the second layer of data blocks stores the ranges of all parameters in the second growth stage, and so on. Each layer of data blocks represents an independent sub-search space, and the order relationship between the layers represents the time progress of the growth period and the progressive order of the optimization calculation. This multi-layered parameter value range data structure organized in time sequence is then used to construct the sequentially associated stage-specific parameter optimization search space.

[0036] Secondly, within each growth stage, the fertilization amount, concentration, and timing parameters of the growth stage constitute the stage decision variables, and the optimization results of the previous growth stage are used as the parameter constraints for the current growth stage. The continuous transfer of parameters between stages can be achieved in the following way: For the target growth stage to be optimized, the feasible value ranges of the three parameters, fertilization amount, concentration, and timing, are extracted from the corresponding level of the search space. These three parameters are defined as decision variables that can be processed by an optimization algorithm, and their value range is strictly limited to the extracted range. Together, they constitute the set of decision variables for the current stage. If the current stage is not the first stage, the optimal combination of fertilization parameters obtained from the previous stage (e.g., the specific fertilization amount value) is used as a known quantity. Its value is used as an initial condition input into the fertilization effect model of the current stage, or directly constrains the starting point of the values ​​of some decision variables in the current stage. Finally, the parameters to be optimized in the current stage are defined as decision variables, coupled with the optimal results from the previous stage as known conditions, thus constituting the stage decision variable definition for achieving continuous information transfer between stages.

[0037] Furthermore, for the current growth stage, under the constraints of the fertilization model, the stage decision variables are combined and solved to construct a stage optimization objective characterizing the crop nutrient demand matching degree and fertilizer utilization efficiency at this stage. This can be achieved in the following way: The stage decision variables of the current stage can be combined and input into the fertilization model. The model can be run to simulate and obtain the predicted cumulative crop nutrient uptake at the end of this stage. The crop nutrient demand matching degree index can then be calculated, that is: based on the theoretical nutrient demand of the crop at this stage, the uptake predicted by the model and the demand can be calculated. The ratio of quantities; calculate the fertilizer utilization efficiency index, that is: divide the cumulative nutrient absorption predicted by the model by the total nutrient input defined by the decision variables in this stage; then use the linear weighting method to integrate the two objectives to form the stage optimization objective function: f(x)=α×P+β×E, where x is the stage decision variable, P is the nutrient demand matching degree, E is the fertilizer utilization efficiency, and α and β are weight coefficients, which are adjusted according to the growth stage, such as α=0.6 and β=0.4 in the vegetative growth stage and α=0.7 and β=0.3 in the fruiting stage, satisfying α+β=1.

[0038] Then, following the chronological order of crop growth stages, parameter optimization calculations are performed stage by stage. After completing the optimization of the current stage, the optimal parameter combination obtained can be used as the fixed boundary condition for subsequent growth stages. This can be achieved in the following way: Starting from the first growth stage, an optimization algorithm is invoked. In this embodiment, a genetic algorithm or sequential quadratic programming can be used. Within the range of decision variable values ​​for the current stage, a specific parameter combination that maximizes the objective function value of the current stage is searched. After optimization, this set of parameters is recorded as the optimal fertilization parameter combination for the current stage. Before entering the optimization calculation for the next growth stage, the specific values ​​of the optimal parameter combination obtained in the previous stage (such as the actual amount of fertilizer) are set as fixed values. These fixed values ​​will serve as the initial soil nutrient conditions or prerequisites for the operation of the fertilization model in subsequent stages. The optimization of decision variables in subsequent stages must be carried out on this basis. Finally, the optimal parameter values ​​obtained from the current stage optimization are fixed and passed to all subsequent growth stages as unchangeable input conditions, thereby achieving a stage-by-stage, conditional optimal sequence solution.

[0039] Finally, after optimization of all growth stages is completed, the optimal fertilization parameter combinations obtained from each stage are integrated to obtain the optimal fertilization parameter combinations covering the entire crop growth period. This can be achieved in the following way: After the optimization calculation of the last growth stage is completed, the optimal fertilization parameter combinations recorded by independent optimization of each stage from the first to the last growth stage are collected. Then, these stage-specific parameter combinations are arranged and assembled according to the natural time sequence of the crop growth period to form a complete and structured fertilization scheme dataset. This dataset clearly specifies the specific fertilization amount, concentration, and timing for each growth stage throughout the entire growth period. Finally, the complete fertilization scheme dataset formed by integrating the independent optimal schemes of each stage in chronological order serves as the stage-specific optimal fertilization parameter combination covering the entire crop growth period and directly guiding precision fertilization operations.

[0040] In step 104, corresponding fertilization control parameters are generated based on the optimal fertilization parameter combination, and the fertilization control parameters are dynamically corrected based on feedback information of crop nutrient status during the fertilization process.

[0041] In some embodiments, generating corresponding fertilization control parameters based on the optimal fertilization parameter combination can be achieved through the following steps: The optimal combination of fertilization parameters is parsed into corresponding fertilization execution parameters; Generate fertilization control instructions based on fertilization execution parameters; The fertilization control commands are configured in a timing sequence to form a fertilization control parameter set; The set of fertilization control parameters is sent to the control unit of the fertilization device.

[0042] It should be noted that the control unit of the fertilization device in this application refers to an embedded hardware controller or programmable logic controller that can receive, parse and execute the fertilization control parameter set, thereby driving the coordinated operation of the various physical components of the fertilization device.

[0043] In practical applications, firstly, the optimal fertilization parameter combination can be parsed into corresponding fertilization execution parameters in the following way: Read the specific parameter values ​​corresponding to the current operation stage from the phased optimal fertilization parameter combination, including the optimal fertilization amount, optimal fertilization concentration, and optimal fertilization timing. Then, based on the technical specifications of the fertilization device, convert these agronomic parameters into direct numerical commands to drive the equipment. For example, parse the fertilization amount into the total stroke or total number of pulses the fertilization pump needs to complete, parse the fertilization concentration into the flow ratio setting value of the mother liquor pump and the clean water pump, and parse the fertilization timing into the fertilization start time and duration based on the irrigation start time. Secondly, the fertilization control command can be generated based on the fertilization execution parameters in the following way: Encode the fertilization execution parameters into specific equipment control commands according to the communication protocol and command format agreed upon with the fertilization device control unit. For example, generate a command containing a target speed value to be sent to the fertilization pump driver. The process involves generating a quantitative fertilizer injection command for the total stroke, or a proportional adjustment command containing the target opening degree sent to the electric valve controller. Then, the timing configuration of the fertilizer control command to form a fertilizer control parameter set can be achieved as follows: Based on the logical order defined by the fertilizer timing parameters, all generated independent control commands are arranged into an ordered task list. In this list, each command is configured with its precise trigger time or trigger condition, thereby ensuring that each device can work collaboratively in a predetermined order during irrigation and fertilization operations. Finally, the fertilizer control parameter set is sent to the control unit of the fertilization device in the following way: Through a communication interface supported by the control unit (such as an RS-485 bus), the configured fertilizer control parameter set is sent as a complete data packet or segmented command sequence to the memory of the control unit. The control unit receives and stores this parameter set, preparing to execute it sequentially when the irrigation operation is triggered.

[0044] In some embodiments, dynamically adjusting the fertilization control parameters based on feedback information about crop nutrient status during fertilization can be achieved through the following steps: During the fertilization process, sensors collect information on the nutrient status of crops. The nutrient status information is compared and analyzed with the pre-fertilized state, and the status deviation is calculated. If the state deviation exceeds a preset tolerance threshold, a control parameter correction mechanism is triggered; Based on the magnitude and direction of the state deviation, one or more parameters in the fertilization control parameters are fine-tuned according to preset correction rules, and subsequent control commands are updated.

[0045] It should be noted that the control parameter correction mechanism in this application refers to the logical process or program module that automatically starts adjusting the fertilization control parameters according to preset rules when the state deviation exceeds the preset tolerance threshold.

[0046] In practical applications, firstly, the collection of crop nutrient status information through sensors during fertilization can be achieved in the following way: one or more chlorophyll fluorescence sensors or canopy multispectral sensors can be used to automatically collect leaf or canopy spectral data of the crop at a preset cycle. Then, through a preset calibration model, the collected raw spectral data is converted into specific values ​​representing leaf nitrogen content or relative chlorophyll content. Secondly, the nutrient status information is compared and analyzed with the pre-fertilization status to calculate the status deviation. This can be achieved in the following way: the theoretical value of crop nutrient status predicted at this moment based on the current fertilization scheme and fertilization action model is read from the system, i.e., the pre-fertilization status. Then, the theoretical value is directly compared with the measured value of real-time nutrient status information, and the absolute difference or relative percentage between the measured value and the theoretical value is calculated. This difference is the status deviation. Then, if the status deviation exceeds a preset tolerance threshold, the control parameter correction mechanism is triggered. The above method is implemented as follows: The absolute value of the calculated state deviation is compared with a tolerance threshold set in advance according to the crop variety and growth stage, allowing for fluctuations. If the absolute value of the state deviation is greater than the tolerance threshold, the system determines that the current nutrient status has significantly deviated from the expectation and automatically activates a correction process for adjusting fertilization control parameters. Finally, based on the magnitude and direction of the state deviation, one or more parameters in the fertilization control parameters are fine-tuned according to preset correction rules, and subsequent control instructions are updated. This can be achieved in the following way: The corresponding adjustment strategy can be found in a preset correction rule table according to the sign and magnitude of the state deviation. For example, if the deviation indication is significantly low, the table is consulted to find an adjustment of increasing the subsequent unexecuted fertilization amount instruction by 5%. Based on this rule, the system automatically updates the values ​​of the set of unexecuted fertilization control parameters stored in the control unit and regenerates the subsequent sequence of control instructions to be executed.

[0047] In step 105, based on the revised fertilization control parameters, liquid fertilizer is applied to the crop root zone simultaneously with irrigation water through the fertilization device, thereby achieving precise fertilization with water and fertilizer synergy that matches the crop growth stage.

[0048] In some embodiments, the following steps can be used to control the application of liquid fertilizer synchronously with irrigation water to the crop root zone via a fertilization device, based on the modified fertilization control parameters: The dosage and timing of liquid fertilizer application should be controlled according to the revised fertilization control parameters. Liquid fertilizer is delivered simultaneously with irrigation water, and the fertilizer-water mixture is guided to the crop root zone through a fertilization device to complete the fertilization process for the corresponding growth stage.

[0049] In practical applications, firstly, controlling the amount and timing of liquid fertilizer application based on the revised fertilization control parameters can be achieved in the following way: The control unit reads and executes the revised fertilization control parameter set, and according to the precise total amount and schedule specified therein, sends a quantitative control command with the target flow rate and total running time to the fertilizer pump driver, while simultaneously sending a timing command for coordinated opening to the main control valve of the irrigation system to ensure that the fertilization action is carried out as planned during the irrigation water flow. Then, the liquid fertilizer and irrigation water are transported synchronously, and the fertilizer-water mixture is guided to the crop root zone through the fertilization device to complete the fertilization execution process for the corresponding growth stage, which can be achieved in the following way: The activated fertilizer pump draws out the liquid fertilizer quantitatively from the storage tank and pumps it through the injection pipeline to the hydraulic or electric fertilizer applicator interface on the main irrigation pipeline. Here, the fertilizer flow mixes instantly with the flowing irrigation water, and the resulting fertilizer-water mixture is then evenly transported and distributed to the root zone soil of the target crop through the drip irrigation network or micro-irrigation tape laid in the field until the preset total amount is completed, after which the system automatically stops.

[0050] On the other hand, in some embodiments, this application provides an intelligent integrated water and fertilizer application device, referencing Figure 4 The figure is a schematic diagram of the structure of an intelligent integrated water and fertilizer application device according to some embodiments of this application. The intelligent integrated water and fertilizer application device 400 includes: a model construction module 401, a feature processing module 402, and a fertilizer application execution module 403, which are described below: Model building module 401, in this application, is mainly used to build a fertilization effect model that describes the relationship between fertilization parameters and crop nutrient absorption. Feature processing module 402, in this application, is used to perform local sensitivity analysis on key fertilization parameters such as fertilization amount, fertilization concentration and fertilization timing according to the fertilization action model, based on the crop growth stage, to identify fertilization parameters that have a significant impact on crop nutrient absorption at each growth stage, and to limit the feasible value range of fertilization parameters at each growth stage according to the analysis results. In this application, the feature processing module 402 is also used to construct a parameter optimization search space with the feasible value range of each growth stage, take the staged fertilization parameter combination as the decision variable, take the comprehensive index of crop nutrient demand matching degree and fertilizer utilization efficiency as the optimization target, and perform parameter optimization calculation under the constraints of the fertilization action model to obtain the optimal fertilization parameter combination. In this application, the feature processing module 402 is also used to generate corresponding fertilization control parameters based on the optimal fertilization parameter combination, and to dynamically correct the fertilization control parameters based on feedback information of crop nutrient status during the fertilization process; The fertilization execution module 403 in this application is mainly used to control the application of liquid fertilizer to the crop root zone simultaneously with irrigation water according to the modified fertilization control parameters, thereby achieving precise water and fertilizer synergy fertilization that matches the crop growth stage.

[0051] In addition, this application also provides a computer device, which includes a memory and a processor. The memory stores code, and the processor is configured to acquire the code and execute the above-described intelligent water and fertilizer integration fertilization parameter optimization method.

[0052] In some embodiments, reference Figure 5 The figure is a schematic diagram of the structure of a computer device for implementing an intelligent water and fertilizer integration fertilization parameter optimization method according to some embodiments of this application. The intelligent water and fertilizer integration fertilization parameter optimization method in the above embodiments can be achieved through... Figure 5 The computer device shown is used to implement this, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0053] Processor 501 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0054] The communication bus 502 can be used to transmit information between the aforementioned components.

[0055] Memory 503 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CDROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 503 may exist independently and be connected to processor 501 via communication bus 502. Memory 503 may also be integrated with processor 501.

[0056] The memory 503 stores program code for executing the scheme of this application, and its execution is controlled by the processor 501. The processor 501 executes the program code stored in the memory 503. The program code may include one or more software modules. In the above embodiment, the intelligent water and fertilizer integration fertilization parameter optimization method can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0057] Communication interface 504 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0058] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single CPU) processor or a multi-core (multi CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0059] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0060] In addition, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described intelligent water and fertilizer integration fertilization parameter optimization method.

[0061] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0062] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for optimizing intelligent fertigation parameters, applied to an intelligent fertigation device, for optimizing crop fertilization parameters, characterized in that... The method includes the following steps: Construct a fertilization effect model to describe the relationship between fertilization parameters and crop nutrient absorption; Based on the fertilization action model, local sensitivity analysis was performed on key fertilization parameters such as fertilization amount, fertilization concentration and fertilization timing at different crop growth stages. The fertilization parameters that significantly affect crop nutrient absorption at each growth stage were identified, and the feasible value range of the fertilization parameters at each growth stage was limited according to the analysis results. A parameter optimization search space is constructed based on the feasible value ranges of each growth stage. The combination of fertilization parameters in stages is used as the decision variable, and the comprehensive index of crop nutrient demand matching degree and fertilizer utilization efficiency is used as the optimization objective. Under the constraints of the fertilization action model, parameter optimization calculation is performed to obtain the optimal combination of fertilization parameters. The corresponding fertilization control parameters are generated based on the optimal fertilization parameter combination, and the fertilization control parameters are dynamically corrected based on the feedback information of crop nutrient status during the fertilization process. Based on the revised fertilization control parameters, liquid fertilizer is applied to the crop root zone simultaneously with irrigation water through a fertilization device, thereby achieving precise fertilization with coordinated water and fertilizer application that matches the crop's growth stage.

2. The method as described in claim 1, characterized in that, Constructing a fertilization action model to describe the relationship between fertilization parameters and crop nutrient uptake specifically includes: Based on the migration and transformation process of nutrients in the rhizosphere soil and the absorption mechanism of crop roots, a mathematical model is established to show the relationship between fertilization parameters and changes in rhizosphere soil nutrient concentration. Crop growth stage and environmental factors are used as regulatory variables in the mathematical model. Values ​​are assigned to the kinetic parameters characterizing nutrient absorption in the model to express the dynamic response relationship as the regulatory variables change. By using historical fertilization experiment data, the parameters to be determined in the mathematical model are calibrated and verified, and the verified mathematical model is used as the fertilization effect model.

3. The method as described in claim 1, characterized in that, Based on the aforementioned fertilization model, local sensitivity analysis was performed on key fertilization parameters such as fertilization amount, concentration, and timing at different crop growth stages. The specific fertilization parameters that significantly affect crop nutrient uptake at each growth stage include: During each predefined crop growth stage, non-analytical fertilization parameters and environmental conditions are kept constant; The individual fertilization parameters of fertilizer application rate, fertilizer concentration, and fertilization timing are perturbed respectively; Calculate the changes in crop nutrient uptake output by the fertilization model under different perturbation conditions; Based on the magnitude of changes in crop nutrient uptake, the sensitivity of each fertilization parameter was ranked, and the parameters with the highest ranking were identified as key parameters that significantly affect nutrient uptake at the crop's growth stage.

4. The method as described in claim 1, characterized in that, Based on the analysis results, the feasible range of fertilization parameters for each growth stage specifically includes: Sensitivity analysis results of key fertilization parameters at each growth stage were obtained; Based on the results of the sensitivity analysis, the permissible range of fertilization parameter changes during each growth stage was determined; By defining the boundaries of the allowable range, feasible value ranges for fertilization parameters at each growth stage are obtained; The feasible value ranges of fertilization parameters for each growth stage are stored as a set of stage-specific fertilization parameter constraints.

5. The method as described in claim 1, characterized in that, A parameter optimization search space is constructed using feasible value ranges for each growth stage. The combination of fertilization parameters applied in stages is used as the decision variable, and the comprehensive index of crop nutrient requirement matching degree and fertilizer utilization efficiency is used as the optimization objective. Parameter optimization calculations are performed under the constraints of the fertilization action model to obtain the optimal fertilization parameter combination, which specifically includes: The feasible value ranges of fertilization parameters for each growth stage are obtained and hierarchically organized according to the time sequence of the growth stages to construct a sequentially related stage parameter optimization search space. Within each growth stage, the fertilization amount, concentration, and timing parameters of the growth stage constitute the stage decision variables, and the optimization results of the previous growth stage are used as the parameter constraints of the current growth stage to achieve continuous transfer of parameters between stages. For the current growth stage, under the constraints of the fertilization model, the stage decision variables are combined and solved to construct a stage optimization objective that characterizes the matching degree of crop nutrient demand and fertilizer utilization efficiency at this stage. Parameter optimization calculations are performed stage by stage according to the order of crop growth stages, and the optimal parameter combination obtained after the current stage optimization is completed is used as the fixed boundary condition for the subsequent growth stages. After optimization of all growth stages, the optimal fertilization parameter combinations obtained at each stage are integrated to obtain the optimal fertilization parameter combinations for each stage covering the entire growth period of the crop.

6. The method as described in claim 1, characterized in that, The specific fertilization control parameters generated based on the optimal fertilization parameter combination include: The optimal combination of fertilization parameters is parsed into corresponding fertilization execution parameters; Generate fertilization control instructions based on fertilization execution parameters; The fertilization control commands are configured in a timing sequence to form a fertilization control parameter set; The set of fertilization control parameters is sent to the control unit of the fertilization device.

7. The method as described in claim 1, characterized in that, Based on the revised fertilization control parameters, the liquid fertilizer is applied synchronously to the crop root zone with irrigation water via a fertilization device, specifically including: The dosage and timing of liquid fertilizer application should be controlled according to the revised fertilization control parameters. Liquid fertilizer is delivered simultaneously with irrigation water, and the fertilizer-water mixture is guided to the crop root zone through a fertilization device to complete the fertilization process for the corresponding growth stage.

8. An intelligent integrated water and fertilizer application device, wherein the fertilization parameters are optimized using the method described in any one of claims 1 to 7, characterized in that, The intelligent integrated water and fertilizer application device includes: The model building module is used to build a fertilization effect model that describes the relationship between fertilization parameters and crop nutrient absorption. The feature processing module is used to perform local sensitivity analysis on key fertilization parameters such as fertilization amount, fertilization concentration and fertilization timing according to the fertilization action model, and to identify fertilization parameters that have a significant impact on crop nutrient absorption at each growth stage. Based on the analysis results, the module limits the feasible value range of fertilization parameters at each growth stage. The feature processing module is also used to construct a parameter optimization search space with the feasible value range of each growth stage, take the staged fertilization parameter combination as the decision variable, take the comprehensive index of crop nutrient demand matching degree and fertilizer utilization efficiency as the optimization target, and perform parameter optimization calculation under the constraints of the fertilization action model to obtain the optimal fertilization parameter combination. The feature processing module is also used to generate corresponding fertilization control parameters based on the optimal fertilization parameter combination, and to dynamically correct the fertilization control parameters based on feedback information of crop nutrient status during the fertilization process. The fertilization execution module is used to control the application of liquid fertilizer to the crop root zone simultaneously with irrigation water, based on the revised fertilization control parameters, thereby achieving precise fertilization of water and fertilizer in accordance with the crop growth stage.

9. A computer device comprising a memory and a processor, the memory storing code, characterized in that, The processor is configured to acquire the code and execute the intelligent water and fertilizer integration fertilization parameter optimization method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the intelligent water and fertilizer integration fertilization parameter optimization method as described in any one of claims 1 to 7.

Images