An intelligent irrigation regulation method and system based on a water-fertilizer coupling model

Abstract

The application discloses an intelligent irrigation regulation and control method and system based on a water and fertilizer coupling model, and relates to the field of intelligent irrigation.The method comprises the following steps: setting a sensor to collect basic data of a planting area, cleaning and data fusing the collected data to obtain a cleaned low-rank matrix and an initial state function; defining a PDE system by using the obtained initial state function, and constructing a water and fertilizer coupling model by introducing crop parameters; determining optimal parameters by using residual calculation, correcting a preliminary model according to the optimal parameters and field collected data, and dynamically adjusting the state; establishing a model scoring mechanism, evaluating the corrected water and fertilizer coupling model in terms of prediction accuracy, and retaining a final model; and based on the final water and fertilizer coupling model, designing an intelligent irrigation strategy by predicting future soil state and crop demand. Through accurate modeling of water and fertilizer dynamics, adaptive regulation and control of optimized parameters, prediction accuracy and intelligent level are improved, and artificial intervention and resource waste are reduced.

Description

Technical Field

[0001] This invention relates to the field of intelligent irrigation, specifically to an intelligent irrigation control method and system based on a water-fertilizer coupling model. Background Technology

[0002] In recent years, smart irrigation systems based on the Internet of Things (IoT), sensor networks, and data-driven approaches have been gradually applied in agriculture. Utilizing multi-sensor networks, environmental parameters such as soil moisture, temperature, and nutrient concentration can be monitored in real time, providing data support for water and fertilizer regulation. Simultaneously, combining mathematical modeling and optimization algorithms enables precise water and fertilizer control, improving water and fertilizer utilization efficiency. However, current smart irrigation systems still face the following technical bottlenecks:

[0003] Data uncertainty and noise issues arise because the complexity of the soil environment leads to noise and missing values ​​in the data collected by sensors. Effective data cleaning to improve data quality is a key issue in optimizing smart irrigation systems.

[0004] The lack of accurate dynamic models for water-nutrient coupling means that existing studies are mostly based on empirical formulas or single physical models, failing to fully couple the transport mechanisms of water and nutrients in soil, resulting in significant biases in predictions. Constructing a mathematical model that accurately describes the dynamic changes in water-nutrient coupling is of great significance for optimizing water and fertilizer management.

[0005] The uncertainty of model parameters and the challenge of dynamic correction: Due to the complexity of factors such as soil characteristics and crop growth conditions, the parameters of water-fertilizer coupling models are difficult to determine accurately. How to optimize parameters by combining field monitoring data and dynamically correct the model to make it more adaptable is an urgent problem to be solved.

[0006] The optimization of intelligent irrigation decision-making strategies is crucial. Existing intelligent irrigation strategies are mostly based on fixed rules or simple feedback control, lacking accurate prediction of future soil conditions, resulting in lagging regulation and difficulty in adapting to real-time environmental changes. Therefore, there is an urgent need for intelligent irrigation strategies based on predictive control to improve the system's adaptability.

[0007] Rational management of soil moisture and nutrients is a crucial aspect of agricultural production, directly impacting crop growth, yield, and resource utilization efficiency. Traditional irrigation and fertilization methods often rely on experience-based decisions, failing to adequately consider the real-time physiological needs of crops, environmental factors, and soil heterogeneity, leading to water waste, nutrient loss, and environmental pollution. Therefore, developing precise and intelligent water and fertilizer regulation technologies is essential for the sustainable development of modern agriculture. Summary of the Invention

[0008] In view of the shortcomings of the prior art described above, the purpose of this invention is to provide an intelligent irrigation control method and system based on a water-fertilizer coupling model to solve the above-mentioned technical problems.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a smart irrigation control method based on a water-fertilizer coupling model, comprising:

[0010] The sensor is set up to collect basic data of the planting area. The collected data is cleaned and fused to obtain a cleaned low-rank matrix and an initial state function.

[0011] The PDE system is defined using the obtained initial state function, and a water-fertilizer coupling model is constructed with the introduced crop absorption and growth parameters.

[0012] The optimal parameters are determined by residual calculation, and the preliminary model is corrected based on the optimal parameters and the field-collected data, and the state is dynamically adjusted.

[0013] Establish a model scoring mechanism to evaluate the prediction accuracy of the corrected water-fertilizer coupling model and retain the final model.

[0014] Based on the final water and fertilizer coupling model, intelligent irrigation strategies are designed by predicting future soil conditions and crop needs.

[0015] The present invention is further configured to obtain a low-rank matrix by cleaning the basic dataset collected by the sensor using a low-rank sparse decomposition method, and to construct an initial water distribution function and an initial nutrient concentration function; the data cleaning calculation logic is as follows: The constraint is X = L + E, where α is the balance coefficient, X is the base dataset, L is a low-rank matrix, E is a sparse noise matrix, and ||L|| * Let ||E||1 be the sum of all singular values ​​of the matrix, and let ||E||1 be the sum of all elements of the matrix, representing the L1 norm. μ0(x) is the initial water distribution function, η0(x) is the initial nutrient concentration function, x is the spatial location, x∈Ω, and Ω is the crop planting area.

[0016] The present invention is further configured to construct a PDE system using partial differential equations: in, The equation for water distribution is as follows: For the nutrient concentration equation, D μ D is the water diffusion parameter. η Γ is a nutrient diffusion parameter, φ and ψ are parameters for crop absorption of water and nutrients, and Γ is a nutrient diffusion parameter. μ and Γ η These refer to the amounts of irrigation and fertilizer input from external sources. and The absorption effect mapping function, For gradient operators, For the moisture distribution gradient, Let μ be the nutrient concentration gradient, μ be the water distribution, and η be the nutrient concentration.

[0017] The present invention is further configured to construct a water-fertilizer coupling model M = {PDE system, θ} by combining the PDE system with the initial state function and the parameters to be optimized. (2) ,μ0(x),η0(x)}, where θ (2) The set of parameters to be optimized includes water diffusion parameters, nutrient diffusion parameters, crop absorption parameters, and external input parameters.

[0018] The present invention is further configured to calculate residuals by combining on-site monitoring data, construct an objective function, and then obtain the optimal parameter set through a global optimization algorithm to correct the water-fertilizer coupling model and state; residual calculation logic: Among them, R μ For moisture residual, R η For nutrient residuals; Objective function calculation logic: J(θ) (2) )=||R(θ (2) )||+λ||μ-μ m || 2 +λ||η-η m || 2 , where J(θ) (2) Optimize the objective function, R(θ) (2) ) represents the residual function of the combined water and nutrient content, λ is the balance factor, and μ is the residual function. m η m This is for on-site monitoring of moisture distribution and nutrient concentration.

[0019] The present invention is further configured such that, by adjusting the objective function J(θ) (2) The solution θ is obtained * , where θ * The optimal parameters are used to correct the moisture state as follows: μ cal (x,t), nutrient state: η cal (x,t); Construct a correction model M using the corrected parameters. * ={PDE system,θ * ,μ cal (x,t),η cal (x,t)}.

[0020] The present invention is further configured with the following accuracy scoring calculation logic: Retention condition S model >B, where S model To determine prediction accuracy, N is the total number of measurement points, and x... iB represents the measurement point location, and B represents the set threshold.

[0021] The present invention is further configured to construct a state prediction operator, obtain the optimal control input using the difference between the predicted value and the retained optimal distribution, and drive the irrigation and fertilization system to reach the optimal state based on the optimal control input; the optimal control input calculation logic is as follows: Among them, v * (x,t) represents the optimal control input, v represents the control input, and J represents the control input. c (v) is the control objective function, and its calculation logic is as follows: Where H(μ,η,v;θ) * ) represents the constructed state prediction operator, T(x,t) represents the target state, r represents the nonlinearity parameter of the target loss, and [t,t+T] represents the target state. p [T] represents the time integration region. p To optimize the control time range, ∫ Ω dx is the integral over the space x. Integrating over time t, where γ is the regularization coefficient. To control the gradient of the input.

[0022] This invention also provides an intelligent irrigation control system based on a water-fertilizer coupling model, the system comprising:

[0023] Data cleaning module: Set up sensors to collect basic data of the planting area, clean the collected data and fuse the data to obtain a cleaned low-rank matrix and an initial state function;

[0024] Model building module: Define the PDE system using the obtained initial state function, and build a water and fertilizer coupling model with the introduced crop absorption and growth parameters;

[0025] Correction and optimization module: Determines the optimal parameters using residual calculation, corrects the preliminary model based on the optimal parameters and field-collected data, and dynamically adjusts the state;

[0026] Model evaluation module: Establish a model scoring mechanism to evaluate the prediction accuracy of the corrected water-fertilizer coupling model and retain the final model;

[0027] Strategy design module: Based on the final water and fertilizer coupling model, intelligent irrigation strategies are designed by predicting future soil conditions and crop needs.

[0028] This invention provides an intelligent irrigation control method and system based on a water-fertilizer coupling model. The method involves collecting basic data from the planting area using sensors, cleaning and fusing the collected data to obtain a cleaned low-rank matrix and an initial state function. A PDE system is defined using the obtained initial state function, and a water-fertilizer coupling model is constructed with introduced crop absorption and growth parameters. Optimal parameters are determined using residual calculations, and the preliminary model is corrected based on the optimal parameters and field-collected data, with dynamic adjustments to the state. A model scoring mechanism is established to evaluate the prediction accuracy of the corrected water-fertilizer coupling model, retaining the final model. Based on the final water-fertilizer coupling model, intelligent irrigation strategies are designed by predicting future soil conditions and crop needs. The beneficial effects include:

[0029] Build accurate water and fertilizer coupling models to improve prediction accuracy.

[0030] By introducing coupled partial differential equations to describe the spatiotemporal evolution of soil moisture and nutrients, and combining crop growth and absorption parameters, this invention achieves dynamic simulation of water and fertilizer migration and crop absorption processes. Compared with traditional empirical formula models, this invention can more accurately predict future soil water and fertilizer status and improve the decision-making accuracy of irrigation and fertilization.

[0031] Optimize model parameters to improve system adaptability

[0032] The residual calculation method is used to construct the optimization objective function, and the key parameters of the water-fertilizer coupling model are dynamically adjusted through the global optimization algorithm. A scoring mechanism is introduced to automatically score the calibration model and select the best model, so that it can adapt to different soil conditions, crop types and environmental changes, and improve the generalization ability and adaptability of the model.

[0033] Improve the level of intelligence and reduce the cost of manual intervention.

[0034] By combining automatic sensor data acquisition, intelligent modeling, adaptive optimization, and predictive control technologies, this invention can achieve fully automatic water and fertilizer regulation, significantly reduce the need for manual intervention, lower agricultural production costs, improve the level of agricultural intelligence, and provide advanced technical support for the development of smart agriculture.

[0035] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0037] Figure 1 A flowchart illustrating an intelligent irrigation control method based on a water-fertilizer coupling model, as an exemplary embodiment of the present invention;

[0038] Figure 2 This is a schematic diagram of the structure of an intelligent irrigation control system based on a water-fertilizer coupling model, which is an exemplary embodiment of the present invention. Detailed Implementation

[0039] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0040] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0041] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0042] Example 1

[0043] A smart irrigation control method based on a water-fertilizer coupling model, such as Figure 1 As shown, it includes:

[0044] The sensor is set up to collect basic data of the planting area. The collected data is cleaned and fused to obtain a cleaned low-rank matrix and an initial state function.

[0045] The PDE system is defined using the obtained initial state function, and a water-fertilizer coupling model is constructed with the introduced crop absorption and growth parameters.

[0046] The optimal parameters are determined by residual calculation, and the preliminary model is corrected based on the optimal parameters and the field-collected data, and the state is dynamically adjusted.

[0047] Establish a model scoring mechanism to evaluate the prediction accuracy of the corrected water-fertilizer coupling model and retain the final model.

[0048] Based on the final water and fertilizer coupling model, intelligent irrigation strategies are designed by predicting future soil conditions and crop needs.

[0049] The present invention is further configured to obtain a low-rank matrix by cleaning the basic dataset collected by the sensor using a low-rank sparse decomposition method, and to construct an initial water distribution function and an initial nutrient concentration function; the data cleaning calculation logic is as follows: The constraint is X = L + E, where α is the balance coefficient, X is the base dataset, L is a low-rank matrix, E is a sparse noise matrix, and ||L|| * Let |E||1 be the L1 norm representing the sum of all singular values ​​of the matrix; let μ0(x) be the initial water distribution function; let η0(x) be the initial nutrient concentration function; let x be the spatial location, x∈Ω, and Ω be the crop planting area. Specifically, the basic data collected by the sensor contains noise and outliers, which need to be processed. A low-rank sparse decomposition method is used to split it into a usable low-rank matrix and a useless sparse noise matrix. The resulting low-rank matrix is ​​then optimized so that the final output is used to construct the initial state function. The initial water distribution function μ0(x) represents the spatial distribution of water within the region, and the initial nutrient concentration function η0(x) represents the spatial distribution within the region. This method effectively reduces data noise, improves model prediction accuracy, and enhances the system's adaptability.

[0050] The present invention is further configured to construct a PDE system using partial differential equations: in, The equation for water distribution is as follows: For the nutrient concentration equation, D μ D is the water diffusion parameter. η Γ is a nutrient diffusion parameter, φ and ψ are parameters for crop absorption of water and nutrients, and Γ is a nutrient diffusion parameter. μ and Γ η These refer to the amounts of irrigation and fertilizer input from external sources. and The absorption effect mapping function, For gradient operators, For the moisture distribution gradient, Let μ be the nutrient concentration gradient, μ be the water distribution, and η be the nutrient concentration. Specifically, the PDE system is a coupled system of partial differential equations. The soil moisture distribution equation describes the evolution of soil moisture over time and spatial location. This represents the diffusion of water in the soil, expressed by the diffusion coefficient D. μ control, This indicates that soil moisture is absorbed by crops, and the absorption rate is affected by water content μ and nutrient concentration η, +Γ μ (x,t) represents external water inputs such as irrigation and rainfall; Nutrient concentration equations represent the evolution of soil nutrients over time and space. This indicates the diffusion of nutrients in the soil. This indicates that nutrients have been absorbed by the crop, +Γ η (x,t) represents the external nutrient input from fertilization; the water diffusion coefficient D μ The value range is 10 -6 ~10 -2 m 2 / s, nutrient diffusion coefficient D η The value range is 10 -8 ~10 -4 m 2 / s, the values ​​of water absorption parameter φ and nutrient absorption parameter ψ range from 0 to 1, and the absorption effect mapping function and The value of Γ is influenced by experience, and the external input Γ μ and Γ η The value ranges from 0 to 1. The PDE system accurately simulates the dynamic changes in soil moisture and nutrients, improving prediction accuracy and resource utilization efficiency.

[0051] The present invention is further configured to construct a water-fertilizer coupling model M = {PDE system, θ} by combining the PDE system with the initial state function and the parameters to be optimized. (2) ,μ0(x),η0(x)}, where θ (2) The set of parameters to be optimized includes water diffusion parameters, nutrient diffusion parameters, crop uptake parameters, and external input parameters. Specifically, the PDE system is a partial differential equation system used to describe the dynamic changes of water and nutrients in the soil. The initial state function represents the water and nutrient distribution state of the soil at the initial moment, determined by field measurements. The set of parameters to be optimized represents the set of parameters that need to be optimized to adjust the PDE model to better reflect actual conditions.

[0052] The present invention is further configured to calculate residuals by combining on-site monitoring data, construct an objective function, and then obtain the optimal parameter set through a global optimization algorithm to correct the water-fertilizer coupling model and state; residual calculation logic: Among them, R μ For moisture residual, R η For nutrient residuals; Objective function calculation logic: J(θ) (2) )=||R(θ (2) )||+λ||μ-μ m || 2 +λ||η-η m || 2 , where J(θ) (2) Optimize the objective function, R(θ) (2) ) represents the residual function of the combined water and nutrient content, λ is the balance factor, and μ is the residual function. m η m This refers to the on-site monitoring of moisture distribution and nutrient concentration. Specifically, the residual construction logic involves calculating the residual using on-site monitoring data and a mathematical model; that is, the error between the prediction model and the actual data. In the objective function, ||R(θ) (2) || represents the norm of the moisture and nutrient residuals. To ensure the model's ability to fit the physical processes, ||μ-μ m || 2 The moisture distribution error is used to measure the deviation between the moisture distribution predicted by the model and the monitoring data, ||η-η m || 2 Nutrient concentration error is used to measure the deviation between the nutrient distribution predicted by the model and the monitoring data. The balance factor λ is used to adjust the weights of the data fitting term and the physical consistency term, and its value ranges from 10. -3 ~10 2 A global optimization algorithm is used to adjust the model parameters so that the objective function reaches its minimum value, thereby finding the optimal water-fertilizer coupling model parameters and improving the prediction accuracy.

[0053] The present invention is further configured such that, by adjusting the objective function J(θ) (2) The solution θ is obtained * , where θ * The optimal parameters are used to correct the moisture state as follows: μ cal (x,t), nutrient state: η cal (x,t); Construct a correction model M using the corrected parameters. * ={PDE system,θ * ,μ cal (x,t),η cal (x,t)}. Specifically, the optimal parameter θ is solved by optimizing the objective function. * And use the optimal parameters to correct the moisture state μ cal (x,t) and nutrient state η cal(x,t) is used to construct the corrected model, which makes the model more consistent with the actual observation data and can more accurately describe the dynamic process of water-fertilizer coupling.

[0054] The present invention is further configured with the following accuracy scoring calculation logic: Retention condition S model >B, where S model To determine prediction accuracy, N is the total number of measurement points, and x... i Here, B represents the measurement point location, and B is the set threshold. Specifically, the prediction accuracy score is used to calculate the degree of agreement between the model's prediction results and the measured data. A higher score indicates a smaller error. The retention threshold B is the set minimum accuracy requirement to ensure that the model meets practical standards. |μ cal (x i ,t)-μ m (x i ,t)| represents the prediction error between the moisture content calculated by the model and the measured moisture content, |η cal (x i ,t)-η m (x i ,t)| represents the prediction error between the nutrient concentration calculated by the model and the measured value. Calculate the average prediction error of the model across all measurement points. To convert prediction error into a score, the score ranges from (0,1], with smaller errors resulting in higher scores. A threshold B is set to filter qualified models. If S... model >B proves that the error is small and the prediction accuracy meets the requirements. If S model The proof that ≤B has a large error requires model adjustment and re-optimization.

[0055] The present invention is further configured to construct a state prediction operator, obtain the optimal control input using the difference between the predicted value and the retained optimal distribution, and drive the irrigation and fertilization system to reach the optimal state based on the optimal control input; the optimal control input calculation logic is as follows: Among them, v * (x,t) represents the optimal control input, v represents the control input, and J represents the control input. c (v) is the control objective function, and its calculation logic is as follows: Where H(μ,η,v;θ) * ) represents the constructed state prediction operator, T(x,t) represents the target state, r represents the nonlinearity parameter of the target loss, and [t,t+T] represents the target state. p [T] represents the time integration region. p To optimize the control time range, ∫ Ω dx is the integral over the space x. Integrating over time t, where γ is the regularization coefficient. To control the gradient of the input, specifically, the state prediction operator H(μ,η,v; θ) is constructed. * The function is to predict the dynamic evolution of the system under a given control input, ensuring that the predicted value is as close as possible to the target value. The prediction operator is obtained from the PDE system, control input, and correction term; T(x,t) is the target state, representing the desired distribution of water and nutrients, ||H(μ,η,v;θ * )-T(x,t)|| r This represents the error between the predicted state and the target state. r, as a nonlinearity parameter, determines the error measurement method, typically taking the value of 1 or 2. If r = 2, it represents the squared error. The gradient of the control input measures its spatial variation. γ is a regularization coefficient used to balance the error term and the smoothness of the control input, preventing excessively drastic changes in the control input. The purpose is to avoid excessively large abrupt changes in the control input in space, ensure that the system is more stable during optimized control, reduce unnecessary resource waste, predict the system state by constructing a state prediction operator, and use the difference between the predicted value and the target state to calculate the optimal control input to optimize the irrigation and fertilization process and achieve the best distribution of water and nutrients.

[0056] Example 2

[0057] Please see Figure 2 This exemplary intelligent irrigation control system based on a water-fertilizer coupling model includes:

[0058] Data cleaning module: Set up sensors to collect basic data of the planting area, clean the collected data and fuse the data to obtain a cleaned low-rank matrix and an initial state function;

[0059] Model building module: Define the PDE system using the obtained initial state function, and build a water and fertilizer coupling model with the introduced crop absorption and growth parameters;

[0060] Correction and optimization module: Determines the optimal parameters using residual calculation, corrects the preliminary model based on the optimal parameters and field-collected data, and dynamically adjusts the state;

[0061] Model evaluation module: Establish a model scoring mechanism to evaluate the prediction accuracy of the corrected water-fertilizer coupling model and retain the final model;

[0062] Strategy design module: Based on the final water and fertilizer coupling model, intelligent irrigation strategies are designed by predicting future soil conditions and crop needs.

[0063] It should be noted that the intelligent irrigation control system based on the water-fertilizer coupling model provided in the above embodiments and the intelligent irrigation control method based on the water-fertilizer coupling model provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the intelligent irrigation control system based on the water-fertilizer coupling model provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the system can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0064] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0065] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0066] In this application, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or multiple items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0067] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0068] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0069] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0070] In the several embodiments provided in this application, it should be understood that the disclosed system can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0071] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0072] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0073] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0074] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A smart irrigation control method based on a water-fertilizer coupling model, characterized in that, include: The system sets up a basic dataset by collecting data from the planting area using sensors. The collected data is then cleaned and fused to obtain a cleaned low-rank matrix and an initial state function. This includes: cleaning the sensor-collected basic dataset using a low-rank sparse decomposition method to obtain a low-rank matrix; constructing an initial water distribution function and an initial nutrient concentration function; and the data cleaning and calculation logic is as follows: Constraints ,in, For balance coefficient, Based on the dataset, It is a low-rank matrix. It is a sparse noise matrix. The nuclear norm represents the sum of all singular values ​​of a matrix. Let L1 norm represent the sum of all elements of a matrix; Let be the initial moisture distribution function. Let be a function of the initial nutrient concentration. For spatial location, , This area is designated for crop cultivation. The PDE system is defined using the obtained initial state function, and a water-fertilizer coupling model is constructed with the introduced crop absorption and growth parameters. The PDE system is a coupled system of partial differential equations, constructed using these equations: ,in, The equation for water distribution is as follows: The nutrient concentration equation is... For moisture diffusion parameters, For nutrient diffusion parameters, and These are parameters for crop absorption of water and nutrients, respectively. and These refer to the amounts of irrigation and fertilizer input from external sources. and The absorption effect mapping function, For gradient operators, For the moisture distribution gradient, For nutrient concentration gradient, For moisture distribution, Here, t represents nutrient concentration, and t represents time. The optimal parameters are determined using residual calculation. Based on these optimal parameters and field-collected data, the preliminary model is calibrated, and the state is dynamically adjusted. This includes: calculating residuals using field monitoring data, constructing an objective function, and then using a global optimization algorithm to obtain the optimal parameter set to calibrate the water-fertilizer coupling model and its state. The residual calculation logic is as follows: ,in, For moisture residual, Nutrient residuals; Objective function calculation logic: ,in, Let be the objective function. This is the residual function of the combined water and nutrient content. As a balance factor, , For on-site monitoring of moisture distribution and nutrient concentration; Establish a model scoring mechanism to evaluate the prediction accuracy of the corrected water-fertilizer coupling model and retain the final model. Based on the final water and fertilizer coupling model, intelligent irrigation strategies are designed by predicting future soil conditions and crop needs.

2. The intelligent irrigation control method based on a water-fertilizer coupling model according to claim 1, characterized in that, Combination Construct a water-fertilizer coupling model with the initial state function and the parameters to be optimized. ,in, The set of parameters to be optimized includes water diffusion parameters, nutrient diffusion parameters, crop absorption parameters, and external input parameters.

3. The intelligent irrigation control method based on a water-fertilizer coupling model according to claim 1, characterized in that, By analyzing the objective function The solution ,in, The optimal parameters are used to correct the moisture state as follows: The nutrient status is as follows: ; Construct a calibration model using the corrected parameters .

4. The intelligent irrigation control method based on a water-fertilizer coupling model according to claim 3, characterized in that, Accuracy score calculation logic: Reservation conditions ,in, To improve prediction accuracy, The total number of measurement points. B represents the measurement point location, and B represents the set threshold.

5. The intelligent irrigation control method based on a water-fertilizer coupling model according to claim 4, characterized in that, A state prediction operator is constructed, and the optimal control input is obtained by using the difference between the predicted value and the retained optimal distribution. The irrigation and fertilization system is then driven to reach its optimal state based on this optimal control input. The optimal control input calculation logic is as follows: ,in, Here, v represents the optimal control input. To control the objective function, the calculation logic is as follows: ,in, For the constructed state prediction operator, For the target state, The nonlinearity parameter of the target loss. For the time integration region, To optimize the control time range, For space Accumulate points. Regarding time Accumulate points. The regularization coefficient is . To control the gradient of the input.

6. A smart irrigation control system based on a water-fertilizer coupling model, used to implement the smart irrigation control method based on a water-fertilizer coupling model as described in any one of claims 1-5, characterized in that, include: Data cleaning module: Set up sensors to collect basic data of the planting area, clean the collected data and fuse the data to obtain a cleaned low-rank matrix and an initial state function; Model building module: Define the PDE system using the obtained initial state function, and build a water and fertilizer coupling model with the introduced crop absorption and growth parameters; Correction and optimization module: Determines the optimal parameters using residual calculation, corrects the preliminary model based on the optimal parameters and field-collected data, and dynamically adjusts the state; Model evaluation module: Establish a model scoring mechanism to evaluate the prediction accuracy of the corrected water-fertilizer coupling model and retain the final model; Strategy design module: Based on the final water and fertilizer coupling model, intelligent irrigation strategies are designed by predicting future soil conditions and crop needs.

Images