Intelligent decision-making system and method for autumn and winter irrigation soil salinity leaching

Abstract

The invention discloses an autumn and winter irrigation-oriented soil salinity leaching intelligent decision-making system and method, and relates to the technical field of intelligent agriculture, the system comprises a double-ring infiltration module, a multi-parameter monitoring column module, a data acquisition module, a decision-making parameter module and an irrigation cloud decision-making module, the double-ring infiltration module is used for monitoring field in-situ infiltration amount data under a non-evaporation condition in real time; the multi-parameter monitoring column module is used for monitoring multi-parameter data of a soil profile in the infiltration process in real time; the data acquisition module is used for acquiring field in-situ infiltration data and multi-parameter data; the decision parameter module is used for acquiring decision parameters input by a user; the irrigation cloud decision-making module is used for making an irrigation decision according to the field in-situ infiltration data, the multi-parameter data and the decision-making parameters and determining the optimal irrigation amount. According to the application, in-situ acquisition of soil water and salt migration key parameters and optimization of autumn and winter irrigation volume can be realized, the stability of salt leaching effect is guaranteed, and the accuracy and efficiency of autumn and winter soil salt leaching decision are improved.

Description

Technical Field

[0001] This application relates to the field of smart agriculture technology, and in particular to a smart decision-making system and method for soil salinity leaching during autumn and winter irrigation. Background Technology

[0002] Autumn and winter irrigation for salt leaching is a key traditional agronomic measure in arid and semi-arid regions of northern my country. It involves a strategic large-scale flood irrigation to leach salts before the ground freezes in late autumn and early winter. Traditional autumn and winter irrigation decisions rely heavily on experience, which presents core challenges such as unpredictable soil salt leaching effects and inaccurate irrigation quotas. The root causes lie in a lack of perception, missing models, and ineffective decision-making.

[0003] In the field of precise water and salt regulation in farmland, integrating dynamic observation of soil water-heat-salt with a dynamic process model of water-heat-salt is a forward-looking approach. However, the reliability of this method is affected by the model structure, boundary conditions, and the accuracy of soil hydraulic and solute transport parameters. Current parameter acquisition methods have significant limitations: traditional indoor measurement methods are inefficient and lack representativeness; some in-situ instruments can only measure saturated hydraulic conductivity, lacking reliable solutions for in-situ measurement of water characteristic curves and unsaturated solute transport parameters. Furthermore, there is an inherent contradiction between the practical needs of field management and existing technological capabilities: fixed irrigation regimes are difficult to adapt to interannual environmental changes, failing to guarantee the stability of salt leaching effects; existing soil parameter acquisition technologies struggle to provide flexible, portable, and representative in-situ data tailored to the differences in soil properties across fields. In summary, traditional autumn and winter irrigation decision-making methods have not yet achieved effective integration among the three key stages of "in-situ perception - model simulation - intelligent decision-making." Various problems in model structure, boundary conditions, and the accuracy of soil hydraulic and solute transport parameters ultimately lead to inaccurate and inefficient autumn and winter soil salt leaching decisions.

[0004] Therefore, there is an urgent need for a smart decision-making system for soil salt leaching during autumn and winter irrigation. This system would enable in-situ acquisition of key parameters of soil water and salt transport and optimization of autumn and winter irrigation amounts, ensuring the stability of salt leaching effects, improving the accuracy and efficiency of autumn and winter soil salt leaching decisions, and solving the problems of unpredictable soil salt leaching effects and inaccurate irrigation quotas caused by reliance on experience in traditional autumn and winter irrigation decision-making methods. This system would also promote the evolution of water and salt management from an experience-based model to a precise and controllable digital model. Summary of the Invention

[0005] The purpose of this application is to provide a smart decision-making system and method for soil salinity leaching in autumn and winter irrigation, which can realize in-situ acquisition of key parameters of soil water and salt transport and optimization of autumn and winter irrigation volume, ensure the stability of salt leaching effect, and improve the accuracy and efficiency of autumn and winter soil salinity leaching decision-making.

[0006] To achieve the above objectives, this application provides the following solution.

[0007] In a first aspect, this application provides a smart decision-making system for soil salinity leaching during autumn and winter irrigation, which includes the following modules.

[0008] A dual-ring infiltration module is installed in the target field area to monitor in-situ infiltration data in the target field area under non-evaporation conditions in real time.

[0009] A multi-parameter monitoring column module is installed in the target field area to monitor multi-parameter data of the soil profile during the infiltration process in the target field area in real time; the multi-parameter data includes: soil moisture content, temperature and electrical conductivity.

[0010] The data acquisition module is connected to the dual-ring infiltration module, the multi-parameter monitoring column module, and the irrigation cloud decision module, respectively, and is used to collect the field in-situ infiltration data and the multi-parameter data and transmit them to the irrigation cloud decision module.

[0011] The decision parameter module, connected to the irrigation cloud decision module, is used to acquire the decision parameters input by the user and transmit them to the irrigation cloud decision module; the decision parameters include: soil salinity before autumn and winter irrigation, crop salt tolerance threshold, and climate conditions.

[0012] The irrigation cloud decision module is used to make irrigation decisions based on the field in-situ infiltration data, the multi-parameter data, and the decision parameters, and to determine the optimal irrigation amount.

[0013] Optionally, the dual-ring infiltration module includes: a dual-ring assembly, a self-recording Mascher bottle, and an outer ring cap.

[0014] The dual-ring assembly includes an inner ring and an outer ring, both of which are partially inserted into the soil of the target field area. The inner ring is located inside the outer ring, and an outer ring cap is provided on the top of the outer ring. The inner ring is used to delineate the test area for infiltration tests under non-evaporation conditions within the test area. The outer ring is used to isolate the test area delineated by the inner ring, ensuring that the aqueous solution inside the inner ring always infiltrates vertically during the infiltration test, reducing lateral seepage. The outer ring cap is used to seal the top of the outer ring, ensuring that water does not evaporate during the infiltration test.

[0015] The outlet of the self-recording Marshall bottle is connected to the inner ring via a flexible hose, and the self-recording Marshall bottle is also electrically connected to the data acquisition module.

[0016] The self-recording Marshall bottle is used to inject aqueous solution into the inner ring through the tubing during the infiltration test, and to record the test time and infiltration volume in real time. The infiltration rate is calculated based on the test time and infiltration volume, and the test time, infiltration volume and infiltration rate are used as the in-situ infiltration volume data in the field and transmitted to the data acquisition module.

[0017] Optionally, the multi-parameter monitoring column module includes: a monitoring column.

[0018] The monitoring column is vertically installed inside the soil at the center of the inner ring, and several sensor components are evenly arranged on the monitoring column. Each sensor component serves as a monitoring node to monitor the soil moisture content, temperature, and electrical conductivity at different monitoring depths.

[0019] Optionally, the monitoring column is also provided with a water inlet, which is sealed to the water outlet of the self-recording Marshall bottle through a hose to form a water supply pipeline, so that the aqueous solution in the self-recording Marshall bottle is injected into the inner ring through the water supply pipeline.

[0020] Optionally, the monitoring depth of the monitoring column is greater than 50cm, and the number of monitoring nodes is greater than or equal to 5.

[0021] Optionally, the irrigation cloud decision module includes: a parameter calibration unit and a scenario simulation unit.

[0022] The parameter calibration unit is used to automatically adjust and determine the optimal key parameters of soil water and salt transport by using a calibration algorithm based on the in-situ infiltration data and the multi-parameter data in the field and by comparing the model prediction data with the real data collected by the data acquisition module. The model prediction data is the infiltration data and multi-parameter data predicted based on the water-heat-salt transport model.

[0023] The scenario simulation unit is used to automatically calculate the optimal irrigation amount required to reduce soil salinity to a safe threshold before autumn and winter irrigation, based on the optimal soil parameters and the decision parameters, with the goal of achieving a safe soil salinity threshold before crop planting, through reverse simulation.

[0024] Optionally, the calibration algorithm includes one or more of the following: LM algorithm, PSO algorithm, MCMC algorithm, and PINN algorithm.

[0025] Optionally, the scenario simulation unit performs a one-dimensional soil profile water-heat-salt numerical simulation under the background of autumn and winter irrigation during the reverse simulation. The initial conditions of the one-dimensional soil profile water-heat-salt numerical simulation are the known initial water-heat-salt data of the soil profile. The upper boundary water and solute conditions are set as atmospheric boundary conditions and zero solute flux boundary conditions during the simulation period, respectively, and the lower boundary water and solute conditions are free drainage and zero solute flux boundary conditions.

[0026] Optionally, when determining the optimal irrigation amount, the scenario simulation unit integrates the freezing model and weather forecast to select a window period with low soil evaporation and no freezing for irrigation; with the goal of ensuring that salt is leached below the depth of strong capillary water rise, the safe burial depth and corresponding irrigation amount required to block salt return are determined; based on the soil salinity content and target desalination rate before autumn and winter irrigation, the water and salt transport process under different irrigation amounts is simulated to construct a quantitative relationship of "irrigation amount - salt burial depth - desalination rate", and the salt leaching depth, winter irrigation time and seasonal freezing index are dynamically adjusted to determine the optimal irrigation amount.

[0027] Secondly, this application proposes a smart decision-making method for soil salinity leaching during autumn and winter irrigation. The smart decision-making method for soil salinity leaching during autumn and winter irrigation is implemented based on the smart decision-making system for soil salinity leaching during autumn and winter irrigation described in the first aspect. The smart decision-making method for soil salinity leaching during autumn and winter irrigation includes the following steps.

[0028] Real-time monitoring of in-situ infiltration data in the target field area under non-evaporation conditions, as well as multi-parameter data of the soil profile during the infiltration process; the multi-parameter data includes: soil moisture content, temperature and electrical conductivity.

[0029] Collect the in-situ infiltration data in the field and the multi-parameter data.

[0030] Obtain decision parameters input by the user; the decision parameters include: soil salinity before autumn and winter irrigation, crop salt tolerance threshold, and climate conditions.

[0031] Irrigation decisions are made based on the in-situ field infiltration data, the multi-parameter data, and the decision parameters to determine the optimal irrigation amount.

[0032] According to the specific embodiments provided in this application, this application has the following technical effects.

[0033] This application provides a smart decision-making system and method for soil salinity leaching during autumn and winter irrigation. The system includes a dual-ring infiltration module, a multi-parameter monitoring column module, a data acquisition module, a decision parameter module, and an irrigation cloud decision module. The dual-ring infiltration module enables real-time monitoring of in-situ infiltration data in the target field area under non-evaporation conditions. The multi-parameter monitoring column module simultaneously captures real-time data on soil moisture content, temperature, and electrical conductivity during the infiltration process. This achieves in-situ, dynamic, and accurate monitoring of soil infiltration-related data and key water-heat-salt data from the soil profile. The data coverage is comprehensive and reflects the actual field conditions, providing reliable data support for in-situ acquisition of key soil water and salt transport parameters and optimization of autumn and winter irrigation amounts. The data acquisition module establishes an efficient data transmission channel between the dual-ring infiltration module, the multi-parameter monitoring column module, and the irrigation cloud decision module, ensuring timely and accurate transmission of in-situ infiltration data and multi-parameter data to the irrigation cloud decision module. This avoids decision-making biases caused by data transmission delays or loss, ensuring the smooth progress of the decision-making process. Furthermore, the decision parameter module allows users to input key decision parameters such as soil salinity before autumn and winter irrigation, crop salt tolerance threshold, and climatic conditions. This enables irrigation decisions to fully consider the initial soil conditions, crop growth needs, and environmental conditions in the target field, making decisions more targeted and personalized, and avoiding the problem of generalized decisions not matching actual field conditions. In addition, by integrating in-situ field monitoring data with user-input decision parameters through the irrigation cloud decision module, the optimal irrigation amount can be accurately determined, ensuring the stability of salt leaching effects and improving the accuracy and efficiency of autumn and winter soil salt leaching decisions. This solves the problems of unpredictable soil salt leaching effects and inaccurate irrigation quotas caused by the reliance on experience in traditional autumn and winter irrigation decision-making methods. It promotes the evolution of water and salt management from an experience-based model to a precise and controllable digital model, achieving the dual goals of water conservation and salt control, and improving the scientific nature and effectiveness of autumn and winter irrigation operations. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a schematic diagram of the structure of a smart decision-making system for soil salinity leaching during autumn and winter irrigation, provided as an embodiment of this application.

[0036] Figure 2 This is a schematic diagram of the in-situ system layout in the field provided in an embodiment of this application.

[0037] Figure 3This is a schematic diagram showing the observed and simulated moisture contents of the topsoil and subsoil layers during the calibration period, as provided in an embodiment of this application.

[0038] Figure 4 A schematic diagram showing the observed and simulated moisture contents of the topsoil and subsoil layers during the verification period, as provided in an embodiment of this application.

[0039] Figure 5 This is a schematic diagram of the observed EC (Electrical Conductivity) and simulated EC of the topsoil and subsoil layers during the calibration period, provided for an embodiment of this application.

[0040] Figure 6 This is a schematic diagram of the observed EC and simulated EC during the verification period of the topsoil and subsoil layers, provided in an embodiment of this application.

[0041] Figure 7 This is a schematic diagram illustrating the prediction of optimal irrigation amount under the scenarios of salt content and entropy return / salt return before autumn and winter irrigation, provided in an embodiment of this application.

[0042] Figure 8 This is a flowchart illustrating a smart decision-making method for soil salinity leaching during autumn and winter irrigation, provided as an embodiment of this application.

[0043] Attached reference numerals: 1-Outer ring cap; 2-Double ring assembly; 3-Multi-parameter monitoring column module; 4-Data acquisition module; 5-Irrigation cloud decision module; 6-Self-recording Maslow bottle. Detailed Implementation

[0044] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0045] The purpose of this application is to provide a smart decision-making system and method for soil salinity leaching during autumn and winter irrigation. This system integrates portable in-situ monitoring of soil profile water-heat-salt during the double-loop infiltration process in the field, a dynamic process model of water-heat-salt, and a cloud-based decision-making module. Based on in-situ, dynamic, and accurate monitoring data of infiltration volume and soil water-heat-salt parameters, it enables in-situ acquisition of key soil parameters and optimization of autumn and winter irrigation amounts. This solves the problems of unpredictable soil salinity leaching effects and inaccurate irrigation quotas caused by the difficulty in obtaining in-situ soil parameters and the complexity of field management in traditional autumn and winter irrigation decision-making methods. It improves the accuracy and efficiency of autumn and winter soil salinity leaching decision-making and promotes the evolution of water and salt management from an experience-based model to a precise and controllable digital model.

[0046] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, this application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0047] like Figure 1 As shown in the figure, this embodiment proposes a smart decision-making system for soil salinity leaching during autumn and winter irrigation. The smart decision-making system for soil salinity leaching during autumn and winter irrigation includes a dual-ring infiltration module, a multi-parameter monitoring column module 3, a data acquisition module 4, a decision parameter module, and an irrigation cloud decision-making module 5.

[0048] The input terminal of the data acquisition module 4 is connected to the output terminal of the dual-ring infiltration module and the output terminal of the multi-parameter monitoring column module 3, respectively. The output terminal of the data acquisition module 4 is connected to the input terminal of the irrigation cloud decision module 5, and the input terminal of the irrigation cloud decision module 5 is also connected to the output terminal of the decision parameter module.

[0049] The dual-ring infiltration module is installed in the target field area to monitor the in-situ infiltration data of the target field area under non-evaporation conditions in real time.

[0050] The multi-parameter monitoring column module 3 is set in the target field area to monitor the multi-parameter data of the soil profile during the infiltration process in the target field area in real time; the multi-parameter data includes: soil moisture content, temperature and electrical conductivity.

[0051] The data acquisition module 4 is connected to the dual-ring infiltration module, the multi-parameter monitoring column module 3, and the irrigation cloud decision module 5, respectively, and is used to collect the field in-situ infiltration data and the multi-parameter data and transmit them to the irrigation cloud decision module 5.

[0052] The decision parameter module is connected to the irrigation cloud decision module 5 and is used to acquire the decision parameters input by the user and transmit them to the irrigation cloud decision module 5. The decision parameters include: soil salinity before autumn and winter irrigation, crop salt tolerance threshold and climate conditions.

[0053] The irrigation cloud decision module 5 is used to make irrigation decisions based on the field in-situ infiltration data, the multi-parameter data, and the decision parameters, and to determine the optimal irrigation amount.

[0054] As an optional implementation, the dual-ring infiltration module includes: a dual-ring assembly 2, a self-recording Maslow bottle 6, and an outer ring cap 1.

[0055] like Figure 2As shown, the dual-ring assembly 2 includes an inner ring and an outer ring, both of which are partially inserted into the soil of the target field area. The inner ring is located inside the outer ring, and the outer ring cap 1 is provided on the top of the outer ring. The inner ring is used to delineate the test area for infiltration tests under non-evaporation conditions. The outer ring is used to isolate the test area delineated by the inner ring, ensuring that the aqueous solution inside the inner ring always infiltrates vertically during the infiltration test, reducing lateral seepage. The outer ring cap 1 is used to seal the top of the outer ring, ensuring that no water evaporates during the infiltration test.

[0056] The outlet of the self-recording Marshall bottle 6 is connected to the inner ring via a flexible hose, and the self-recording Marshall bottle 6 is also electrically connected to the data acquisition module 4.

[0057] The self-recording Marshall bottle 6 is used to inject aqueous solution into the inner ring through the hose during the infiltration test, and to record the test time and infiltration volume in real time. The infiltration rate is calculated based on the test time and infiltration volume, and the test time, infiltration volume and infiltration rate are used as the field in-situ infiltration volume data and transmitted to the data acquisition module 4.

[0058] As an optional implementation, the multi-parameter monitoring column module 3 includes: a monitoring column.

[0059] The monitoring column is vertically installed inside the soil at the center of the inner ring, and several sensor components are evenly arranged on the monitoring column. Each sensor component serves as a monitoring node to monitor the soil moisture content, temperature, and electrical conductivity at different monitoring depths.

[0060] As an optional implementation, the monitoring column is also provided with a water inlet, which is sealed to the water outlet of the self-recording Marshall bottle 6 through a hose to form a water supply pipeline, so that the aqueous solution in the self-recording Marshall bottle 6 is injected into the inner ring through the water supply pipeline.

[0061] As an optional implementation, the monitoring depth of the monitoring column is greater than 50cm, and the number of monitoring nodes is greater than or equal to 5.

[0062] As an optional implementation, the irrigation cloud decision module 5 includes: a parameter calibration unit and a scenario simulation unit.

[0063] The parameter calibration unit is used to automatically adjust and determine the optimal key parameters of soil water and salt transport by using a calibration algorithm based on the field in-situ infiltration data and the multi-parameter data, by comparing the model prediction data with the real data collected by the data acquisition module 4; the model prediction data is the infiltration data and multi-parameter data predicted based on the water-heat-salt transport model.

[0064] The scenario simulation unit is used to automatically calculate the optimal irrigation amount required to reduce soil salinity to a safe threshold before autumn and winter irrigation, based on the optimal soil parameters and the decision parameters, with the goal of achieving a safe soil salinity threshold before crop planting, through reverse simulation.

[0065] As an optional implementation, the calibration algorithm includes one or more of the following: LM (Levenberg-Marquardt) algorithm, PSO (Particle Swarm Optimization) algorithm, MCMC (Markov Chain Monte Carlo) algorithm, and PINN (Physics-Informed Neural Networks) algorithm.

[0066] As an optional implementation, the scenario simulation unit performs a one-dimensional soil profile water-heat-salt numerical simulation under the background of autumn and winter irrigation during the back-dive simulation. The initial conditions of the one-dimensional soil profile water-heat-salt numerical simulation are the known initial water-heat-salt data of the soil profile. The upper boundary water and solute conditions are set as atmospheric boundary conditions and zero solute flux boundary conditions during the simulation period, respectively, and the lower boundary water and solute conditions are free drainage and zero solute flux boundary conditions.

[0067] As an optional implementation, when determining the optimal irrigation amount, the scenario simulation unit integrates the freezing model and weather forecast to select a window period with low soil evaporation and no freezing for irrigation; with the goal of ensuring that salt is leached below the depth of strong capillary water rise, the safe burial depth and corresponding irrigation amount required to block salt return are determined; based on the soil salinity content and target desalination rate before autumn and winter irrigation, the water and salt transport process under different irrigation amounts is simulated to construct a quantitative relationship of "irrigation amount - salt burial depth - desalination rate", and the salt leaching depth, winter irrigation time and seasonal freezing index are dynamically adjusted to determine the optimal irrigation amount.

[0068] To make the technical solution of this application clearer, the system structure and specific implementation of this application will be described in detail below with examples.

[0069] This embodiment proposes a smart decision-making system for soil salinity leaching in autumn and winter irrigation. The system includes a dual-ring infiltration module, a multi-parameter monitoring column module 3, a data acquisition module 4, a decision parameter module, and an irrigation cloud decision module 5.

[0070] The dual-ring infiltration module, comprising a dual-ring component 2, a self-recording Marshall bottle 6, and an outer ring cap 1, is used for in-situ field infiltration and infiltration volume monitoring under non-evaporation conditions. The dual-ring component 2 consists of an inner ring and an outer ring, which are two concentric ring structures. The inner ring delineates the test area (the soil area within the inner ring is considered the test area), thereby controlling the test area. The outer ring ensures vertical infiltration of the aqueous solution from the inner ring, thus reducing lateral seepage.

[0071] The multi-parameter monitoring column module 3 is located at the center of the double-ring component 2 and inserted into the soil. It monitors various parameters such as soil moisture content, temperature, and salinity. It is mainly used for in-situ monitoring of soil moisture content, temperature, and electrical conductivity in the soil profile during infiltration. The monitoring depth of the multi-parameter monitoring column module 3 is greater than 50 cm, and the number of monitoring nodes is greater than or equal to 5.

[0072] The data acquisition module 4 includes a data acquisition unit and a wireless transmission unit, which are used for data acquisition and transmission in the dual-ring infiltration module and the multi-parameter monitoring column module 3.

[0073] The decision parameter module includes soil salinity before autumn and winter irrigation, crop salt tolerance threshold, and climate conditions. Users can input these decision parameters to provide a basis for predicting the optimal irrigation amount.

[0074] The irrigation cloud decision-making module 5 includes a parameter calibration unit and a scenario simulation unit. The parameter calibration unit utilizes calibration algorithms such as the LM algorithm and the PSO algorithm to automatically adjust and find the most suitable parameters by comparing model predictions with real data from the data acquisition module 4. The scenario simulation unit aims to achieve a safe soil salinity threshold before crop planting. Through reverse simulation, it automatically calculates the optimal irrigation amount required to reduce soil salinity to the safe threshold before autumn and winter irrigation.

[0075] In this embodiment, in-situ dynamic monitoring of infiltration rate and multiple soil parameters is achieved by coupling a dual-ring infiltration module and a multi-parameter monitoring column module 3. The installation and collaborative operation of the dual-ring infiltration module and the multi-parameter monitoring column module 3 are as follows: a hole is drilled to a set depth using a soil drill, and the monitoring column of the multi-parameter monitoring column module 3 is screwed into the soil to continuously collect dynamic data on soil moisture, temperature, and salinity during the infiltration process, providing a basis for the calibration of key soil parameters. The inner and outer rings of the dual-ring infiltration module are installed around the monitoring column of the multi-parameter monitoring column module 3. A flexible hose is used to tightly connect the outlet of the self-recording Marshall bottle 6 (based on automatic infiltration monitoring) to the inlet of the monitoring column of the multi-parameter monitoring column module 3, ensuring a good seal in the water supply pipeline. During the test, the outer ring cap 1 is used to ensure no significant evaporation at the upper boundary.

[0076] In this embodiment, the testing process includes a controlled infiltration stage and a zero-input drainage stage, which are determined by the infiltration rate and duration, respectively. The controlled infiltration rate is recommended to be around the amount of conventional irrigation. The duration of the zero-input drainage stage is adjusted based on the soil salinity and texture, ideally until the drainage stage parameters stabilize to ensure the data meets the calibration requirements. For sandy loam soil, the recommended drainage stage duration is as follows: approximately 120 hours for low-salinity soils (less than 4 g / kg) and approximately 48 hours for medium-salinity soils (4-8 g / kg). Furthermore, for high-salinity soils (greater than 8 g / kg), salt content significantly interferes with water content data; therefore, pre-irrigation should be conducted to reduce soil salinity before testing. Soil parameter acquisition methods are based on the known boundary conditions of the above stages and water-salt observation data from the monitoring column.

[0077] During the initial to steady-state infiltration stage, the steady-state infiltration rate is automatically calculated using a self-recording Marshall bottle 6 with automatic monitoring. Hydraulic and solute transport parameters are calibrated using data from the controlled infiltration and drainage stages, conducted in two steps: The dynamic changes in soil moisture are simulated using the Richards equation, which describes one-dimensional unsteady flow in the soil, as expressed in the following formula.

[0078] (1).

[0079] in, Volumetric water content (m³) 3 / m 3 ), This is the hydraulic conductivity parameter (m / day). Let m be the matrix potential. z It is in the vertical direction.

[0080] Represented using the Van Genuchten-Mualem model and and The relationship between them can be expressed as the following formula.

[0081] (2).

[0082] (3).

[0083] in, Residual moisture content (m) 3 / m 3 ), Saturated water content (m 3 / m 3 ), The saturated hydraulic conductivity is 1 m / day. For effective saturation, and =( , α , m, n, l These are all empirical shape parameters. m and n The relationship is defined as follows m =1-1 / n ,in, and l They are fixed at 0.05 and -0.5 respectively. and cultivated layer These are measured values, while the parameters... α、n and subsoil Calibration needs to be performed based on measured moisture content and other data.

[0084] In this embodiment, upper and lower boundary conditions are set for different stages. The upper boundary conditions for the infiltration and drainage stages are controlled as water flow rates with known flux (changes in water flux within the self-recording Martens bottle 6) and zero flux, respectively, while the lower boundary condition is free drainage. The soil profile moisture content before the start of the infiltration stage is used as the initial condition. The one-dimensional soil profile is set as a two-layer structure of topsoil and subsoil, and the calculated soil saturated hydraulic conductivity is used. K s1 (Approximately equal to steady-state infiltration rate) and known saturated water content of the topsoil ( ) and subsoil saturated water content ( Calibrate the remaining soil hydraulic parameters, including pore parameters ( α and n ) and the saturated hydraulic conductivity of the subsoil layer ( K s2 ).

[0085] Regarding parameter calibration methods, this application supports multiple optimization algorithms, including but not limited to the classic LM algorithm and the PSO algorithm series, which are suitable for fast local parameter estimation and global optimization, respectively. After obtaining the optimal parameter estimate, the Markov chain Monte Carlo method, as a powerful sampler, can approximate the posterior distribution of the parameters by constructing a Markov chain, thereby accurately quantifying the uncertainty of the parameters. In addition, the Physical Information Neural Network achieves a deep integration of physical mechanisms and data-driven approaches by embedding physical equations as constraints into the loss function. This characteristic enables it to effectively alleviate the dependence on large amounts of high-precision data, providing a new approach for handling challenging scenarios such as sparse data and high noise. This multi-algorithm fusion design concept allows this application to adapt to different application scenarios: from real-time field estimation requiring rapid response, to scientific research requiring accurate quantification of uncertainty, and to parameter identification under sparse data conditions, satisfactory results can be obtained by selecting appropriate calibration methods, providing a comprehensive solution for soil parameter calibration.

[0086] After the hydraulic parameters were calibrated, the solute transport parameters were calibrated by simulating soil salinity changes using the convection-dispersion equation, which describes the one-dimensional solute transport law in soil water. The result is expressed as follows.

[0087] (4).

[0088] in, Solute concentration (kg / m³) 3 ), The Darcy velocity is expressed in m / day. Hydrodynamic dispersion coefficient (m 2 / day), z Vertical direction This indicates adsorption.

[0089] Hydrodynamic dispersion coefficient This includes mechanical dispersion and molecular diffusion. Mechanical dispersion is mainly caused by the action of water, while molecular diffusion mainly depends on the properties of the solute itself and the temperature.

[0090] (5).

[0091] in, The bending factor is typically a function of soil moisture content. The diffusion coefficient (m) of the solute in free water 2 / day), Longitudinal dispersion (m). The average pore flow velocity is (m / day).

[0092] (6).

[0093] in, For the allocation coefficient (m) 3 / kg).

[0094] In this embodiment, boundary conditions are set for different stages. The upper boundary condition for the infiltration and drainage stages is a zero-flux solute flow rate, and the lower boundary condition is a zero-concentration-gradient solute flow rate. The salinity content before the start of the infiltration and drainage stages is used as the initial condition, respectively. Furthermore, the conductivity formula obtained by the multi-parameter monitoring column module 3 needs to be derived from the known apparent conductivity (EC). a ) and pore water conductivity (EC w The relationship between the two factors needs to be transformed, and the effect of temperature changes on conductivity needs to be considered (conductivity decreases by approximately 2% for every 1°C decrease in temperature). Using the above conditions and conductivity data, solute transport parameters are calibrated, including the longitudinal dispersion of solutes in the topsoil and subsoil layers. DL1 , D L2 ) and allocation coefficient ( K d1 , K d2 Molecular diffusion coefficient D w The impact on the simulation results is minimal, so it is set to 0. Furthermore, chemical reactions between salts are not considered in this application; therefore, the chemical reaction rate constant is set to 0.

[0095] (7).

[0096] in, denoted as apparent conductivity (dS / m). The electrical conductivity of pore water is dS / m. and These are the azithromycin cementation index and the azithromycin saturation index, respectively. In this embodiment, and The values ​​are 1.5 and 2 respectively.

[0097] In this embodiment, the specific method for realizing intelligent irrigation decision-making is as follows: A one-dimensional soil profile water-heat-salt numerical simulation is conducted under the background of autumn and winter irrigation. The initial conditions for the simulation are known initial water-heat-salt data of the soil profile. The upper boundary water and solute conditions are set as atmospheric boundary conditions (including daily temperature, precipitation, wind speed, etc.) and zero solute flux boundary conditions during the simulation period, respectively. The lower boundary water and solute conditions are free drainage and zero solute flux boundary conditions. The initial water-heat-salt data of the soil profile, calibration parameters, winter meteorological conditions, and salt leaching targets are used as input decision parameters. Finally, the irrigation cloud decision module 5 dynamically generates irrigation schemes by collaboratively utilizing in-situ monitoring data, the calibrated water-heat-salt transport model, and meteorological and freezing forecasts. This is fundamentally different from existing technologies that rely on experience to fix irrigation regimes or single threshold constraints, achieving a leap from "experience-driven" to "dual-driven by process models and real-time data," and achieving precise water saving while ensuring stable salt leaching effects.

[0098] In the main scenarios corresponding to autumn and winter irrigation, the irrigation cloud decision-making system, by integrating freezing models and weather forecasts, accurately selects the window period for irrigation when soil evaporation is low and the soil is not frozen, thus avoiding the risks of premature evaporation and delayed infiltration. Regarding the issue of spring salt return, the system aims to ensure that salt is leached below the depth of strong capillary water rise, thereby determining the safe burial depth and corresponding irrigation volume required to block salt return. Furthermore, based on the soil salinity content before autumn and winter irrigation and the target desalination rate, the system simulates the water-salt transport process under different irrigation volumes, constructing a quantitative relationship of "irrigation volume - salt burial depth - desalination rate". Therefore, the irrigation cloud decision-making module 5 dynamically adjusts key decision parameters such as salt leaching depth, winter irrigation time, and seasonal freezing index through scenario simulation, thereby achieving water-saving goals while ensuring stable salt leaching effects, and ultimately determining the optimal irrigation volume.

[0099] In this embodiment, a slightly saline-alkali land suitable for corn cultivation was selected as the test site, and the soil profile structure was surveyed using boreholes. The double-ring infiltration module and multi-parameter monitoring column module 3 were placed on a flat plot for infiltration testing, and the data was connected to the data acquisition module 4. After the infiltration test, the outer ring cover 1, the double rings, and the data acquisition unit were left in place for approximately 48 hours to collect data during the drainage phase. After data acquisition, the data was transmitted to the irrigation cloud decision module 5 for parameter calibration and prediction of the optimal irrigation amount.

[0100] In this embodiment, the intelligent decision-making system for soil salinity leaching during autumn and winter irrigation is used to quickly obtain soil parameters and predict winter irrigation amounts. The main steps include the following steps.

[0101] (1) An infiltration test was conducted using the multi-parameter monitoring column module 3 and the double-ring infiltration module to obtain the infiltration rate and dynamic changes in water, heat and salt during the infiltration process.

[0102] (2) Data obtained from the infiltration test (including field in-situ infiltration data and multi-parameter data) are acquired through data acquisition module 4 to perform model parameterization and irrigation volume prediction.

[0103] (3) The key soil parameters (including soil hydraulic parameters and solute transport parameters) are calibrated through the irrigation cloud decision module 5 to obtain the key soil parameters required by the model.

[0104] (4) Based on the input decision parameters, the optimal irrigation amount is determined through the irrigation cloud decision module 5.

[0105] In this embodiment, the calibration results of soil hydraulic parameters and solute transport parameters obtained based on observation data are shown in Table 1.

[0106] Table 1. Calibration results of soil hydraulic parameters and solute transport parameters

[0107] Figure 3 A schematic diagram showing the observed and simulated moisture contents of the topsoil and subsoil layers during the calibration period is presented. Figure 4 A schematic diagram showing the observed and simulated moisture contents of the topsoil and subsoil layers during the verification period is presented. Figure 5 A schematic diagram of the observed EC and simulated EC during the calibration period of the topsoil and subsoil layers is shown. Figure 6 A schematic diagram showing the observed EC and simulated EC during the validation period for the topsoil and subsoil layers is presented. Figure 3 , Figure 4 , Figure 5 and Figure 6 It can be seen that the calibrated parameters can simulate soil moisture content well (calibration period R). 2 =0.80, RMSE=0.02m 3 / m 3 ) and dynamic changes in conductivity (calibration period R) 2 =0.80, RMSE=0.06dS / m). Additionally, when using the calibrated parameters for verification, soil moisture content (verification period R) is also considered. 2 =0.85, RMSE=0.2m 3 / m 3 The accuracy of conductivity simulation has been further improved (validation period R). 2 =0.85, RMSE=0.01dS / m). This verification result proves that the present application can achieve reliable in-situ parameter acquisition and prediction.

[0108] This embodiment uses a cornfield as an example. The decision-making system accepts three types of decision-making information constraints, including a salt tolerance threshold, soil salinity before autumn / winter irrigation, and the irrigation scenario. In this embodiment, the crop type is corn. The irrigation cloud decision module 5 matches the corresponding salt threshold (2.4 g / kg) based on the input crop type. After inputting the decision parameters, the irrigation cloud decision module 5 automatically runs the model and calculates the optimal irrigation amount. The specific results in this embodiment are as follows: Figure 7 As shown. The results indicate that the optimal irrigation amount is influenced by both soil salinity before autumn / winter irrigation and climatic conditions. Within the range of 2.5-6.0 g / kg soil salinity before autumn / winter irrigation, the optimal irrigation amount increases with increasing salinity; and at any salinity level, the increasing pattern follows the order of warm winter salinization < conventional winter salinization < cold winter salinization. Furthermore, 150m... 3 The conventional irrigation amount per mu is only suitable for slightly saline-alkali land, while the optimal irrigation amount (200-250 m³ / mu) is suitable for moderately saline-alkali land. 3 / mu) far exceeds this value; conversely, under normal or warm winter conditions, slightly saline-alkali land only requires 50-120m 3 / mu.

[0109] Based on the same inventive concept, this application also provides a method for intelligent decision-making regarding soil salinity leaching during autumn and winter irrigation, based on the aforementioned intelligent decision-making system for soil salinity leaching during autumn and winter irrigation. The solution provided by this method is similar to the solution described in the aforementioned system. Therefore, the specific limitations in the embodiments of the method for intelligent decision-making regarding soil salinity leaching during autumn and winter irrigation provided below can be found in the limitations of the intelligent decision-making system for soil salinity leaching during autumn and winter irrigation described above, and will not be repeated here.

[0110] In one exemplary embodiment, such as Figure 8 As shown, a smart decision-making method for soil salinity leaching in autumn and winter irrigation is provided, which specifically includes the following steps.

[0111] S1: Real-time monitoring of in-situ infiltration data in the target field area under non-evaporation conditions, as well as multi-parameter data of the soil profile during the infiltration process; the multi-parameter data includes: soil moisture content, temperature, and electrical conductivity.

[0112] S2: Collect the in-situ infiltration data in the field and the multi-parameter data.

[0113] S3: Obtain the decision parameters input by the user; the decision parameters include: soil salinity before autumn and winter irrigation, crop salt tolerance threshold and climate conditions.

[0114] S4: Make irrigation decisions based on the field in-situ infiltration data, the multi-parameter data, and the decision parameters to determine the optimal irrigation amount.

[0115] In practical applications, infiltration tests are first conducted using the multi-parameter monitoring column module 3 and the double-ring infiltration module to obtain data on the infiltration rate and dynamic changes in soil moisture, temperature, and salinity during the infiltration process. Then, the data acquisition module 4 collects and transmits the data from the infiltration test to provide data support for model parameterization and irrigation volume prediction. Next, the parameter calibration module of the irrigation cloud decision module 5 calibrates the soil hydraulic parameters and solute transport parameters based on the collected data to obtain the key soil parameters required by the model. Then, the user inputs the soil salinity before autumn and winter irrigation, the crop salt tolerance threshold, and climatic conditions through the decision parameter module. Finally, the scenario simulation module of the irrigation cloud decision module 5 dynamically generates the optimal irrigation volume and irrigation plan based on the calibrated parameters, the user-input decision parameters, and the one-dimensional soil profile water-heat-salt numerical simulation under the autumn and winter irrigation background.

[0116] The intelligent decision-making system and method for soil salinity leaching in autumn and winter irrigation proposed in this application have the following technical advantages.

[0117] (1) An integrated closed-loop system of "in-situ sensing-model simulation-intelligent decision-making" was constructed, which effectively bridged the gaps in the three key links of the traditional autumn and winter irrigation decision-making method and realized the transformation of water and salt management from an experience-based model to a precise and controllable digital model. Through the collaborative work of the double-loop infiltration module and the multi-parameter monitoring column module 3, in-situ, dynamic and precise monitoring of infiltration rate and soil water-heat-salt parameters was realized, solving the problems of low efficiency, insufficient representativeness and difficulty in in-situ measurement of traditional parameters.

[0118] (2) The irrigation cloud decision module 5 integrates multiple calibration algorithms, which can be flexibly selected according to different application scenarios. It can meet the needs of real-time field estimation with rapid response, and can also adapt to scientific research with precise quantification of uncertainty and parameter identification under data sparse conditions. It provides a comprehensive and reliable solution for the calibration of key soil parameters and ensures the accuracy of model parameters.

[0119] (3) The scenario simulation module is driven by both process models and real-time data, integrates freezing models and weather forecasts, and combines key parameters such as crop salt tolerance thresholds to determine the optimal irrigation amount through reverse simulation, thus realizing dynamic optimization of the irrigation scheme. Compared with the traditional fixed irrigation system that relies on experience, this application can adapt to the differences in environmental factors and soil properties between years, and achieve precise water saving while ensuring the stability of salt leaching effect, significantly improving the scientificity and practicality of autumn and winter irrigation.

[0120] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0121] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. An intelligent decision system for autumn and winter irrigation soil salt leaching, characterized in that, The autumn and winter irrigation soil salt leaching intelligent decision system comprises: A double-ring infiltration module arranged in a target field area and used for monitoring in-situ field infiltration data of the target field area under non-evaporation conditions in real time; A multi-parameter monitoring column module arranged in the target field area and used for monitoring multi-parameter data of a soil profile during infiltration of the target field area in real time; the multi-parameter data comprises soil water content, temperature and conductivity; A data acquisition module connected with the double-ring infiltration module, the multi-parameter monitoring column module and an irrigation cloud decision module respectively, and used for acquiring the in-situ field infiltration data and the multi-parameter data and transmitting them to the irrigation cloud decision module; A decision parameter module connected with the irrigation cloud decision module, and used for acquiring decision parameters input by a user and transmitting them to the irrigation cloud decision module; the decision parameters comprise soil salt content before autumn and winter irrigation, crop salt tolerance threshold and climate conditions; The irrigation cloud decision module is used for making irrigation decisions according to the in-situ field infiltration data, the multi-parameter data and the decision parameters, and determining an optimal irrigation amount.

2. The autumn and winter facing irrigation soil salt leaching intelligent decision system according to claim 1, characterized in that, The double-ring infiltration module comprises a double-ring assembly, a self-recording Mariotte bottle and an outer ring cover. The double-ring assembly comprises an inner ring and an outer ring; the inner ring and the outer ring are both partially inserted into soil of the target field area, and the inner ring is located inside the outer ring; a top portion of the outer ring is provided with the outer ring cover; the inner ring is used for circumscribing a test area, so that an infiltration test under non-evaporation conditions is performed in the test area; the outer ring is used for isolating the test area circumscribed by the inner ring, so that water solution inside the inner ring always vertically infiltrates during the infiltration test, and lateral seepage is reduced; the outer ring cover is used for sealing the top portion of the outer ring, so that water does not evaporate during the infiltration test; A water outlet of the self-recording Mariotte bottle is connected to the inside of the inner ring through a hose, and the self-recording Mariotte bottle is also electrically connected with the data acquisition module; The self-recording Mariotte bottle is used for injecting water solution into the inner ring through the hose during the infiltration test, and recording test time and infiltration amount in real time; an infiltration rate is calculated based on the test time and the infiltration amount; the test time, the infiltration amount and the infiltration rate are taken as the in-situ field infiltration data, and are transmitted to the data acquisition module.

3. The autumn and winter facing irrigation soil salt leaching intelligent decision system according to claim 2, characterized in that, The multi-parameter monitoring column module comprises a monitoring column. The monitoring column is vertically arranged inside soil at the center of the inner ring, and a plurality of sensor assemblies are uniformly arranged on the monitoring column; each sensor assembly is taken as a monitoring node, so as to monitor soil water content, temperature and conductivity at a soil profile at different monitoring depths.

4. The autumn and winter facing irrigation soil salt leaching intelligent decision system according to claim 3, characterized in that, The monitoring column is also provided with a water inlet; the water inlet is sealingly connected with the water outlet of the self-recording Mariotte bottle through a hose, so as to form a water supply pipeline; water solution in the self-recording Mariotte bottle is injected into the inside of the inner ring through the water supply pipeline.

5. The autumn and winter soil irrigation salt leaching intelligent decision system according to claim 3, characterized in that, The monitoring depth of the monitoring column is greater than 50 cm, and the number of the monitoring nodes is greater than or equal to 5.

6. The autumn and winter soil irrigation salt leaching intelligent decision system according to claim 1, characterized in that, The irrigation cloud decision module comprises a parameter calibration unit and a scenario simulation unit. The parameter calibration unit is configured to automatically adjust and determine optimal soil parameters by using a calibration algorithm to compare model prediction data with real data collected by the data collection module based on the field in-situ infiltration amount data and the multi-parameter data; the model prediction data are infiltration amount data and multi-parameter data predicted based on a water-heat-salt transport model; The scenario simulation unit is configured to automatically calculate the optimal irrigation amount required to reduce soil salinity to a safe threshold before autumn and winter irrigation by reverse deduction simulation based on the optimal soil parameters and the decision parameters, with the goal of achieving a safe threshold of soil salinity before crop planting.

7. The autumn and winter facing irrigation soil salt leaching intelligent decision system according to claim 6, characterized in that, The calibration algorithm includes one or more of the LM algorithm, the PSO algorithm, the MCMC algorithm, and the PINN algorithm.

8. The autumn and winter facing irrigation soil salt leaching intelligent decision system according to claim 6, characterized in that, The scenario simulation unit performs one-dimensional soil profile water-heat-salt numerical simulation under the background of autumn and winter irrigation during reverse deduction simulation; the initial conditions of the one-dimensional soil profile water-heat-salt numerical simulation are known soil profile initial water-heat-salt data; wherein the upper boundary water and solute conditions are set as atmospheric boundary conditions and zero solute flux boundary conditions during the simulation period, respectively, and the lower boundary water and solute conditions are free drainage and zero solute flux boundary conditions.

9. The autumn and winter facing soil salinity leaching intelligent decision system according to claim 6, characterized in that, When determining the optimal irrigation amount, the scenario simulation unit fuses a freezing model and weather forecasts to select a window period with low soil evaporation power and no freezing for irrigation; to ensure that the salt is leached to a depth below the capillary water rise depth, the required safe burial depth and the corresponding irrigation amount are determined; based on the soil salinity content before autumn and winter irrigation and the target desalination rate, the water and salt transport process under different irrigation amounts is simulated to construct a quantitative relationship of "irrigation amount-buried salt depth-desalination rate", and the optimal irrigation amount is determined by dynamically adjusting the salt leaching depth, winter irrigation time, and seasonal freezing index.

10. An intelligent decision-making method for autumn and winter irrigation soil salt leaching, characterized in that, The intelligent decision-making method for soil salt leaching during autumn and winter irrigation is implemented based on the intelligent decision-making system for soil salt leaching during autumn and winter irrigation of any one of claims 1-9, and the intelligent decision-making method for soil salt leaching during autumn and winter irrigation comprises: Real-time monitoring of field in-situ infiltration amount data and multi-parameter data of a target field area under non-evaporation conditions; the multi-parameter data include soil moisture content, temperature, and electrical conductivity; Collecting the field in-situ infiltration amount data and the multi-parameter data; Obtaining user input decision parameters; the decision parameters include soil salinity before autumn and winter irrigation, crop salt tolerance threshold, and climate conditions; Making irrigation decisions based on the field in-situ infiltration amount data, the multi-parameter data, and the decision parameters to determine the optimal irrigation amount.

Images