CN-120570205-B - Intelligent irrigation regulation and control method and system based on water-fertilizer coupling model

CN120570205BCN 120570205 BCN120570205 BCN 120570205BCN-120570205-B

Abstract

The invention discloses an intelligent irrigation regulation and control method and system based on a water-fertilizer coupling model, and relates to the field of intelligent irrigation. The method comprises the steps of setting basic data of a planting area acquired by a sensor, cleaning the acquired data, fusing the data to obtain a cleaned low-rank matrix and an initial state function, defining a PDE (PDE) system by using the obtained initial state function, constructing a water-fertilizer coupling model with introduced crop parameters, determining optimal parameters by residual calculation, correcting the initial model according to the optimal parameters and the field acquired data, dynamically adjusting the state, establishing a model scoring mechanism, evaluating the corrected water-fertilizer coupling model in terms of prediction accuracy, reserving to obtain a final model, and designing an intelligent irrigation strategy by predicting future soil states and crop demands based on the final water-fertilizer coupling model. By accurately modeling the water and fertilizer dynamics, optimizing parameter self-adaptive regulation and control, the prediction precision and the intelligent level are improved, and the manual intervention and the resource waste are reduced.

Inventors

SHU XIAOJUAN
Fu Duoduo
YU SHIMIN
WU JUAN
ZHANG JIANGONG
PAN CONG

Assignees

江苏上膳源生态农业发展有限公司

Dates

Publication Date: 20260508
Application Date: 20250623

Claims (6)

1. An intelligent irrigation regulation and control method based on a water-fertilizer coupling model is characterized by comprising the following steps: setting basic data of a planting area collected by a sensor to construct a basic data set, cleaning the collected data and fusing the data to obtain a cleaned low-rank matrix and an initial state function, wherein the method comprises the steps of cleaning the basic data set collected by the sensor by using a low-rank sparse decomposition method to obtain the low-rank matrix, constructing an initial water distribution function and an initial nutrient concentration function, and carrying out data cleaning calculation logic: Constraint conditions , wherein, In order for the coefficient of balance to be present, As a basis for the data set of the data, In the case of a low-rank matrix, In order to have a sparse noise matrix, For the sum of all singular values of the kernel norm representation matrix, Representing the sum of all elements of the matrix for the L1 norm; as a function of the initial moisture distribution, As a function of the initial nutrient concentration, In order to be a spatial location, , Is a crop planting area; defining a PDE system by using the obtained initial state function, and constructing a water-fertilizer coupling model with the introduced crop absorption and growth parameters, wherein the PDE system is a coupling partial differential equation set, and constructing by using a partial differential equation: , wherein, In order to be a water distribution equation, In order to be a nutrient concentration equation, As a parameter of the diffusion of the water, As a parameter of the diffusion of the nutrient, And The absorption parameters of crops on moisture and nutrient are respectively, And Respectively the irrigation and fertilization amounts input from the outside, And In order to absorb the effect mapping function, For the purpose of the gradient operator, In order to provide a gradient of the distribution of water, Is the concentration gradient of the nutrient, For the distribution of the water content, The nutrient concentration is the nutrient concentration, and t is the time; The method comprises the steps of calculating and obtaining residual errors by combining field monitoring data, constructing an objective function, obtaining an optimal parameter set by a global optimization algorithm, correcting a water-fertilizer coupling model and a state, and carrying out residual error calculation logic: , wherein, As a residual of the moisture content, Calculating logic for objective function: , wherein, As a function of the object to be processed, As a residual function of the combined moisture and nutrient, As a balance factor, the balance factor is, , The water distribution and the nutrient concentration are monitored on site; Establishing a model scoring mechanism, evaluating the corrected water-fertilizer coupling model on the prediction accuracy, and reserving to obtain a final model; Based on a final water-fertilizer coupling model, an intelligent irrigation strategy is designed by predicting future soil states and crop demands.
2. The intelligent irrigation regulation and control method based on the water-fertilizer coupling model as claimed in claim 1, wherein the method is combined with the following steps of Constructing a water-fertilizer coupling model with the initial state function and the parameters to be optimized , wherein, The parameter set to be optimized comprises a moisture diffusion parameter, a nutrient diffusion parameter, a crop absorption parameter and an external input parameter.
3. The intelligent irrigation regulation and control method based on the water-fertilizer coupling model according to claim 1, characterized by the fact that by aiming at the objective function Solving to , wherein, As the optimal parameters, the moisture state is corrected by using the optimal parameters as follows: The nutrient state is as follows: constructing a correction model by using the corrected parameters 。
4. The intelligent irrigation regulation and control method based on the water-fertilizer coupling model according to claim 3, wherein accuracy score calculation logic: Reservation condition , wherein, In order to predict the accuracy of the device, For the total number of measurement points, For measuring the point position, B is a set threshold.
5. The intelligent irrigation regulation and control method based on the water and fertilizer coupling model according to claim 4, wherein a state predictor is constructed, an optimal control input is obtained by utilizing a difference value between the predicted value and the reserved optimal distribution, and an irrigation and fertilization system is driven to reach an optimal state according to the optimal control input, and the optimal control input calculation logic is used for: , wherein, For optimal control inputs, v is the control input, To control the objective function, the calculation logic is: , wherein, In order to construct the state predictors of the state, In order to be in the target state, As a non-linearity parameter of the target loss, In order to integrate the regions in time, In order to optimize the time frame of the control, For the space of The integration is performed and the integration is performed, Time of The integration is performed and the integration is performed, For the regularization coefficient(s), To control the gradient of the input.
6. An intelligent irrigation regulation and control system based on a water-fertilizer coupling model for realizing the intelligent irrigation regulation and control method based on the water-fertilizer coupling model as set forth in any one of claims 1 to 5, comprising: the data cleaning module is used for setting the sensor to collect basic data of the planting area, cleaning the collected data and fusing the data to obtain a cleaned low-rank matrix and an initial state function; the model construction module is used for defining a PDE system by using the obtained initial state function and constructing a water-fertilizer coupling model with the introduced crop absorption and growth parameters; The correction and optimization module is used for determining optimal parameters by residual calculation, correcting the preliminary model according to the optimal parameters and field acquisition data and dynamically adjusting the state; the model evaluation module is used for establishing a model scoring mechanism, evaluating the corrected water-fertilizer coupling model on the prediction accuracy, and reserving to obtain a final model; and the strategy design module is used for designing an intelligent irrigation strategy by predicting the future soil state and the crop demand based on the final water-fertilizer coupling model.

Description

Intelligent irrigation regulation and control method and system based on water-fertilizer coupling model Technical Field The invention relates to the field of intelligent irrigation, in particular to an intelligent irrigation regulation and control method and system based on a water-fertilizer coupling model. Background In recent years, intelligent irrigation systems based on the Internet of things, sensor networks and data driving are gradually applied to the agricultural field. By means of the multi-sensor network, environmental parameters such as soil humidity, temperature, nutrient concentration and the like can be monitored in real time, and data support is provided for water and fertilizer regulation. Meanwhile, by combining mathematical modeling and an optimization algorithm, the water and fertilizer can be accurately regulated and controlled, and the water and fertilizer utilization efficiency is improved. However, current intelligent irrigation systems still have the following technical bottlenecks: Data uncertainty and noise problems, due to the complexity of the soil environment, noise and missing values exist in the data acquired by the sensor. How to effectively clean the data and improve the data quality is a key problem of intelligent irrigation system optimization. The lack of an accurate water and fertilizer coupling dynamic model, the existing research is mostly based on an empirical formula or a single physical model, and the transmission mechanism of moisture and nutrients in soil cannot be fully coupled, so that the deviation of a prediction result is larger. The construction of a mathematical model capable of accurately describing the dynamic change of the water-nutrient coupling has important significance for optimizing the water and fertilizer management. The problems of uncertainty of model parameters and dynamic correction are solved, and the parameters of the water-fertilizer coupling model are difficult to accurately determine due to the complexity of factors such as soil characteristics, crop growth conditions and the like. How to combine on-site monitoring data to perform parameter optimization and dynamically correct a model so that the model has stronger adaptability is a problem to be solved urgently. The optimization of the intelligent irrigation decision strategy is based on fixed rules or simple feedback control, and the existing intelligent irrigation strategy lacks accurate prediction of future soil states, so that regulation and control lag is caused, and the real-time environment change is difficult to adapt. Therefore, there is a need for intelligent irrigation strategies based on predictive control that improve the adaptive capabilities of the system. Reasonable management of soil moisture and nutrients is a key link in agricultural production, and directly affects crop growth, yield and resource utilization efficiency. Traditional irrigation and fertilization methods often depend on empirical decisions, and cannot fully consider the problems of real-time physiological demands of crops, environmental factors and heterogeneity of soil, such as water resource waste, nutrient loss, environmental pollution and the like. Therefore, the development of accurate and intelligent water and fertilizer regulation and control technology is important for the sustainable development of modern agriculture. Disclosure of Invention Based on the defects of the prior art, the invention aims to provide an intelligent irrigation regulation and control method and system based on a water-fertilizer coupling model so as to solve the technical problems. In order to achieve the purpose, the invention provides the following technical scheme that the intelligent irrigation regulation and control method based on the water-fertilizer coupling model comprises the following steps: Setting a sensor to collect basic data of a planting area, and cleaning and 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-fertilizer coupling model with the introduced crop absorption and growth parameters; Determining optimal parameters by residual calculation, correcting the preliminary model according to the optimal parameters and field acquisition data, and dynamically adjusting the state; Establishing a model scoring mechanism, evaluating the corrected water-fertilizer coupling model on the prediction accuracy, and reserving to obtain a final model; Based on a final water-fertilizer coupling model, an intelligent irrigation strategy is designed by predicting future soil states and crop demands. The invention is further configured to obtain a low-rank matrix by cleaning a basic data set acquired by a sensor by using a low-rank sparse decomposition method, and construct an initial water distribution function and an initial nutrient concentration function, wherein the data cleaning