Intelligent irrigation and soil conditioning system for saline-alkali soil improvement

Abstract

The invention relates to the technical field of saline-alkali soil improvement, in particular to an intelligent irrigation and soil conditioning system for saline-alkali soil improvement, which comprises a climate prediction module, a soil detection module, an electrochemical salt migration prediction module, a data analysis module, an irrigation execution module and a data feedback module, the climate prediction module predicts future climate conditions; the soil detection module monitors soil parameters of the saline-alkali land in real time; the electrochemical salt migration prediction module is used for predicting the migration behavior of salt in soil; the data analysis module formulates irrigation and soil conditioning strategies; the irrigation execution module controls operation of irrigation equipment; and the data feedback module feeds back the irrigation execution result and the actual soil improvement effect, and continuously optimizes irrigation and soil conditioning strategies. According to the method, efficient utilization of resources is achieved, and the accuracy and scientificity of saline-alkali soil improvement are ensured.

Description

Technical Field

[0001] The invention relates to the technical field of saline-alkali land improvement, and in particular to an intelligent irrigation and soil conditioning system for saline-alkali land improvement. Background Art

[0002] The improvement of saline-alkali land has always been an important issue in agricultural production, especially in arid and semi-arid areas. The widespread distribution of saline-alkali land has seriously affected the growth of crops and the health of the soil. Traditional improvement methods mainly rely on large-scale irrigation and application of soil conditioners to reduce the salt concentration in the soil through dilution and chemical modification. However, this method is often inefficient and difficult to achieve precise control, resulting in waste of water resources and unsatisfactory improvement effects. In addition, with the increasing uncertainty brought about by climate change, a single improvement strategy cannot adapt to changing environmental conditions. Therefore, there is an urgent need for an intelligent and dynamic saline-alkali land improvement system to cope with complex natural conditions.

[0003] Existing saline-alkali land improvement technologies mostly rely on manual experience and static irrigation and conditioning plans, which are difficult to dynamically adjust according to real-time environmental changes. Although some technologies have begun to introduce automated monitoring methods, there are still many shortcomings in how to comprehensively analyze the complex relationship between climate, soil and salt migration. Traditional systems lack intelligent prediction and optimization capabilities, cannot achieve precise control of salt migration paths, and are difficult to adjust irrigation and conditioning strategies through real-time data feedback. This not only affects the improvement effect, but may also lead to further deterioration of soil quality and waste of water resources.

[0004] The present invention aims to provide an intelligent irrigation and soil conditioning system for saline-alkali land improvement, which can flexibly adjust irrigation parameters and the application amount of conditioning agents according to actual soil and climatic conditions, effectively control the migration path of salt, thereby improving soil quality and crop yields, optimizing water resource utilization, and ensuring stable and efficient operation of the system in complex environments. Summary of the invention

[0005] The invention provides an intelligent irrigation and soil conditioning system for improving saline-alkali land.

[0006] The intelligent irrigation and soil conditioning system for saline-alkali land improvement includes a climate prediction module, a soil detection module, an electrochemical salt migration prediction module, a data analysis module, an irrigation execution module and a data feedback module, wherein;

[0007] The climate prediction module predicts future climate conditions, including rainfall, temperature, humidity and evaporation, based on real-time climate data and historical climate data;

[0008] The soil detection module monitors the soil parameters of saline-alkali land in real time, including salt concentration, pH value, moisture content and organic matter content;

[0009] The electrochemical salt migration prediction module predicts the migration behavior of salt in the soil based on the soil parameters of saline-alkali land and the predicted climate conditions, specifically including:

[0010] Electric field application: Apply an electric field of predetermined strength and direction in the soil to promote the orderly migration of salt in the soil. According to the monitored soil parameters, the electric field strength and direction are dynamically adjusted to control the movement path of the salt.

[0011] Salt migration monitoring: responsible for real-time monitoring of the concentration changes and migration speed of salt in the soil, and analyzing the distribution of salt at different depths;

[0012] Migration prediction: Combined with the predicted climate conditions, the migration behavior and distribution trend of salt are predicted through the electrochemical salt migration prediction model;

[0013] Feedback correction: According to the real-time monitoring data of salt migration monitoring, the electrochemical salt migration prediction model is corrected and the model parameters are adjusted;

[0014] The data analysis module comprehensively analyzes the predicted results of climate conditions and salt migration behavior and soil parameters, formulates irrigation and soil conditioning strategies, and generates irrigation instructions, including adjusting irrigation amount, irrigation time, irrigation method, and application of conditioning agents;

[0015] The irrigation execution module controls the operation of the irrigation equipment according to the generated irrigation instructions, including the adjustment of water volume, irrigation time and irrigation method;

[0016] The data feedback module provides feedback on the irrigation execution results and the actual soil improvement effects, and continuously optimizes the irrigation and soil conditioning strategies by real-time monitoring of the irrigation and conditioning effects.

[0017] Optionally, the climate prediction module includes:

[0018] Data collection: Obtain historical and current climate data from historical databases and real-time weather stations, including historical and real-time temperature, humidity, rainfall, wind speed, and evaporation;

[0019] Data preprocessing: standardize and denoise the collected climate data;

[0020] Climate prediction model construction: Use multivariate linear regression model to predict climate parameters, expressed as:

[0021] T t+1 =α1T t +α2H t +α3R t +1;

[0022] Among them, T t+1 is the predicted temperature value for the next time step, T t is the temperature value of the current time step, H t is the current humidity value, R t is the current rainfall, α1, α2, α3 are regression coefficients, and 1 is the error term;

[0023] H t+1 =β1H t +β2T t +β3W t +2;

[0024] Among them, H t+1 is the predicted humidity value for the next time step, W t is the current wind speed value, β1, β2, β3 are regression coefficients, and 2 is the error term;

[0025] R t+1 =γ1R t +γ2T t +γ3W t +3;

[0026] Among them, R t+1 is the predicted rainfall value for the next time step, γ1, γ2, γ3 are regression coefficients, and 3 is the error term;

[0027] E t+1 =δ1E t +δ2T t +δ3W t +4;

[0028] Among them, E t+1 is the predicted value of evaporation at the next time step, δ1, δ2, δ3 are regression coefficients, and 4 is the error term;

[0029] Model optimization: Determine regression coefficients using the least squares method, and adjust model parameters through cross-validation and residual analysis;

[0030] Climate Conditions Prediction: Use an optimized multivariate linear regression model to predict climate conditions for future time periods and generate predicted values ​​of temperature, humidity, precipitation, and evaporation for future time steps.

[0031] Optionally, the soil detection module includes:

[0032] Sensor deployment: Multiple soil sensors are deployed in different areas and depths of saline-alkali land to collect real-time data on salt concentration, pH value, moisture content and organic matter content. Soil sensors include conductivity sensors, pH sensors, soil moisture sensors and near-infrared spectrum sensors;

[0033] Data collection: Regularly collect data on soil salt concentration, pH value, moisture content, and organic matter content;

[0034] Data processing: preliminary processing of the collected soil parameters, including data cleaning, denoising and standardization.

[0035] Optionally, the electric field application includes:

[0036] Electric field application: An electric field of predetermined strength and direction is applied through a series of electrodes in the saline-alkali land to form an electric potential difference. The direction θ is determined by the relative positions of the electrodes and is expressed as:

[0037]

[0038] Where V is the voltage between electrodes, and d is the distance between electrodes;

[0039] Salt migration control: Under the action of the electric field, the salt ions in the soil will migrate in an orderly manner along the direction of the electric field to form an electroosmotic flow. According to the monitored salt concentration changes, the salt migration speed v is calculated in real time, expressed as:

[0040] v = μE;

[0041] Where, μ is the electrical mobility of salt ions, and E is the electric field strength;

[0042] Dynamically adjust the electric field: According to real-time soil parameters (such as salt concentration and soil conductivity), the intensity E′ and direction θ′ of the electric field are dynamically adjusted through a feedback control algorithm, which is expressed as:

[0043] E′(t)=E+k1ΔS(t);

[0044] θ′(t)=θ+k2Δσ(t);

[0045] Among them, ΔS(t) is the change of salt concentration monitored in real time, Δσ(t) is the change of soil conductivity, k1 and k2 are adjustment coefficients, and t is time;

[0046] Path control and optimization: By controlling the direction of the electric field, a directional salt migration path is formed, and the direction of the electric field is continuously adjusted using the gradient descent method.

[0047] Optionally, the salt migration monitoring includes:

[0048] Salt concentration monitoring: Conductivity sensors are placed at different soil depths to collect soil conductivity data at each depth in real time, expressed as:

[0049] S(z,t)=a·σ(z,t)+b;

[0050] Where S(z,t) is the salt concentration at time t and depth z, σ(z,t) is the conductivity at the corresponding depth, and a and b are empirical constants;

[0051] Migration velocity calculation: Based on the change in salt concentration at different time points, the migration velocity of salt is calculated and expressed as:

[0052]

[0053] Where v(z,t) is the migration velocity, is the rate of change of salt concentration with time, Δz is the depth interval;

[0054] Salt distribution analysis: The Kriging interpolation method is used to spatially interpolate the salt concentration data at different depths to generate a distribution map of salt in the soil profile, which is expressed as:

[0055]

[0056] in, is the estimated salt concentration at depth z, S(z i ) is the known depth z i The salt concentration at i is the interpolation weight.

[0057] Optionally, the electrochemical salt migration prediction model adopts a time convolutional network (TCN) model, and the time convolutional network (TCN) model includes:

[0058] Input data preprocessing: soil parameters and predicted climate conditions are used as input data and standardized;

[0059] Multi-channel input convolution: Process time series inputs from multiple data channels (such as temperature, humidity, rainfall, soil salt concentration, etc.) and extract the spatiotemporal dependencies of each feature, expressed as:

[0060]

[0061] in, is the output of the kth convolution kernel at time step t, The convolution weight of the kth convolution kernel for the cth channel, is the input data segment (length d) of the cth channel at time step t, and C is the number of channels of the input data;

[0062] Causal convolution: Ensure that the convolution operation only considers the data of the current time step and before, expressed as:

[0063]

[0064] Among them, yt is the convolution output at time step t, W i is the weight of the i-th convolution kernel, x t-i is the input data of time step ti;

[0065] Weighted climate condition convolution layer: The weight parameter is introduced in the convolution layer to emphasize the different effects of different climate conditions (such as temperature, humidity, rainfall) on salt migration, which is expressed as:

[0066]

[0067] Among them, y t is the output at time step t, α j is the weight coefficient corresponding to the jth climate condition, W j is the weight of the jth convolution kernel, x t-j is the input data of time step tj;

[0068] Residual connection and skip connection: Residual connection and skip connection are introduced to enhance the depth of the network and the ability to capture long-term dependencies, expressed as:

[0069] y t ′=y t +x t ;

[0070] Among them, y t ′ is the output after residual connection, y t is the output of the current convolutional layer, x t is the input of the current time step;

[0071] y t ″=y t ′+f(x t-k );

[0072] Among them, y t ″ is the output after skip connection, f(·) is the ReLU activation function;

[0073] Salt migration prediction: Based on the output of multi-layer convolution, the salt migration behavior and distribution trend in the future time step are predicted, expressed as:

[0074]

[0075] Among them, S t+1 is the predicted value of salt migration in the next time step, W m is the weight of the mth prediction layer, y t ″ is the output of the last convolution layer, and M is the total number of prediction layers.

[0076] Optionally, the feedback correction includes:

[0077] Real-time data acquisition: Obtain real-time monitored salt concentration and migration velocity data and compare them with the output of the electrochemical salt migration prediction model;

[0078] Error calculation: Calculate the error between the predicted value and the actual monitored value, expressed as:

[0079] (t) = S actual (t)-S predicted (t);

[0080] Where, (t) is the error value at time t, S actual (t) is the actual monitored salt concentration, S predicted (t) is the salt concentration predicted by the model;

[0081] Parameter adjustment: According to the calculated error, the model parameters are adjusted through the gradient descent algorithm, which is expressed as:

[0082]

[0083] Among them, θ new is the updated model parameter, θ old is the original model parameter, η is the learning rate, is the gradient of the error with respect to the model parameters;

[0084] Model calibration: Apply the adjusted parameters to the electrochemical salt migration prediction model to update the model's predictive capabilities in real time.

[0085] Optionally, the data analysis module includes:

[0086] Data collection and standardization: Collect and standardize data on climate conditions, salt migration predictions, and soil parameters;

[0087] Multidimensional feature fusion: A weighted fusion method is used to integrate the standardized climate conditions, soil parameters, and salt migration prediction results into a comprehensive index for evaluating current soil conditions and future trends, expressed as:

[0088] F = α1·C + α2·S + α3·T;

[0089] Among them, F is the integrated index after fusion, C is the climate condition, S is the salt migration prediction result, T is the soil parameter, and α1, α2, and α3 are the corresponding weight coefficients;

[0090] Irrigation and soil conditioning strategy formulation: Based on the comprehensive index F, the irrigation and soil conditioning strategy is formulated through the optimization algorithm to determine the optimal irrigation amount, irrigation time, irrigation method and application amount of the conditioner. The optimization goal is to maximize crop yield and soil health, which is expressed as:

[0091]

[0092] Among them, G is the objective function, y i is the contribution of the ith decision variable (such as irrigation amount, time, etc.) to the target, λ i is the weight coefficient, c i is the cost, ν i To adjust the parameters;

[0093] Irrigation instruction generation: Generate irrigation instructions based on the optimization results, including adjusting the irrigation amount Q, irrigation time T, irrigation method M and the application amount A of the conditioner, expressed as:

[0094] Q = β1·F + γ1;

[0095] U = β2·F + γ2;

[0096] A=β3·F+γ3;

[0097]

[0098] Among them, Q is the irrigation amount, U is the irrigation time, A is the amount of conditioner applied, M is the irrigation method, β1, β2, β3 are decision coefficients, γ1, γ2, γ3 are adjustment constants, and F1 and F2 are critical values.

[0099] Optionally, the irrigation execution module includes:

[0100] Instruction reception: receiving the generated irrigation instructions, including adjusting the irrigation amount Q, irrigation time T and the application amount A of the conditioner;

[0101] Equipment parameter setting: Automatically adjust the operating parameters of irrigation equipment according to the generated irrigation instructions, including:

[0102] Water volume control: The water volume control of irrigation equipment is based on pump flow rate regulation or valve opening control, expressed as:

[0103] Q = P·t;

[0104] Among them, Q is the water volume set in the irrigation instruction, P is the flow rate (unit is volume / time), and t is the irrigation duration;

[0105] Time setting: The irrigation time is set to the start time of irrigation. start and end time U end , expressed as:

[0106] U end =U start +U;

[0107] Where, U is the duration of irrigation;

[0108] Irrigation mode selection: According to the M value in the irrigation instruction, adjust the working mode of the irrigation equipment, expressed as:

[0109]

[0110] Optionally, the data feedback module includes:

[0111] Real-time monitoring and data collection: The sensor network monitors the soil moisture content, salt concentration, pH value and residual concentration of conditioners during irrigation in real time, and collects relevant data, including the immediate impact of irrigation amount, irrigation time and irrigation method on the soil;

[0112] Error calculation and evaluation: The monitoring data is compared with the expected irrigation effect and soil improvement target, and the error is calculated and expressed as:

[0113] i (t) = S actual,i (t)-S predicted,i (t);

[0114] in, i (t) is the error of the i-th monitoring parameter at time t, S actual,i (t) is the actual monitored value, S predicted,i (t) is the value predicted in the data analysis module;

[0115] Kalman filter correction and adjustment: Apply the Kalman filter algorithm to correct the monitoring data, and adjust the irrigation strategy and soil conditioning strategy based on the corrected monitoring data;

[0116] Strategy optimization and update: Based on the adjusted strategy parameters, the irrigation and soil conditioning strategies are updated and new irrigation instructions are generated.

[0117] Beneficial effects of the present invention:

[0118] The present invention, through the comprehensive use of modules such as climate prediction, soil detection and electrochemical salt migration prediction, can acquire and analyze multi-dimensional data of saline-alkali land in real time, accurately predict climate conditions and salt migration behavior, and combine with the intelligent decision-making of the data analysis module. The system can generate optimal irrigation and conditioning instructions for different soil conditions and climate conditions, realize efficient utilization of resources, and ensure the accuracy and scientificity of saline-alkali land improvement.

[0119] The present invention dynamically adjusts the parameters of the irrigation equipment, including the amount of water, irrigation time and irrigation method, through the generated irrigation instructions, so as to flexibly adapt to different environmental conditions. Through multi-level optimization algorithms and intelligent control means, the system can continuously optimize the irrigation strategy, improve execution efficiency, ensure stable operation in complex and changeable environments, effectively prevent excessive accumulation or loss of salt, and maximize the effect of soil improvement.

[0120] The present invention uses the Kalman filtering algorithm to accurately correct the monitoring data, reduce noise interference, and adjust the irrigation and soil conditioning strategies in real time to ensure that the system can always make accurate responses based on the latest environmental data. Through this precise real-time feedback mechanism, the system can continuously adapt to changes in soil and climate conditions, optimize the improvement process of saline-alkali land, and ensure long-term soil health and crop yield increases. BRIEF DESCRIPTION OF THE DRAWINGS

[0121] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0122] Figure 1 A schematic diagram of system function modules according to an embodiment of the present invention;

[0123] Figure 2 Schematic diagram of an electrochemical salt migration prediction module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0124] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. At the same time, it is explained here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0125] It should be noted that the references to "one embodiment", "an embodiment", "an exemplary embodiment", "some embodiments" and the like in the specification indicate that the embodiments described may include specific features, structures or characteristics, but not every embodiment may include the specific features, structures or characteristics. In addition, when a specific feature, structure or characteristic is described in conjunction with an embodiment, it should be within the knowledge of a person skilled in the art to implement such feature, structure or characteristic in conjunction with other embodiments (whether or not explicitly described).

[0126] In general, a term can be understood, at least in part, from its use in context. For example, depending, at least in part, on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending, at least in part, on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0127] like Figure 1-Figure 2 As shown, the intelligent irrigation and soil conditioning system for saline-alkali land improvement includes a climate prediction module, a soil detection module, an electrochemical salt migration prediction module, a data analysis module, an irrigation execution module and a data feedback module, wherein;

[0128] The climate prediction module predicts future climate conditions, including rainfall, temperature, humidity, and evaporation, based on real-time and historical climate data;

[0129] The soil detection module monitors the soil parameters of saline-alkali land in real time, including salt concentration, pH value, moisture content and organic matter content;

[0130] The electrochemical salt migration prediction module predicts the migration behavior of salt in the soil based on the soil parameters of saline-alkali land and the predicted climate conditions, that is, it analyzes the distribution changes of salt under future climate conditions in real time, including:

[0131] Electric field application: Apply an electric field of predetermined strength and direction in the soil to promote the orderly migration of salt in the soil. According to the monitored soil parameters, the electric field strength and direction are dynamically adjusted to control the movement path of the salt.

[0132] Salt migration monitoring: responsible for real-time monitoring of the concentration changes and migration speed of salt in the soil, and analyzing the distribution of salt at different depths;

[0133] Migration prediction: Combined with the predicted climate conditions, the migration behavior and distribution trend of salt are predicted through the electrochemical salt migration prediction model;

[0134] Feedback correction: According to the real-time monitoring data of salt migration monitoring, the electrochemical salt migration prediction model is corrected, the model parameters are adjusted, and the prediction accuracy is improved;

[0135] The data analysis module conducts a comprehensive analysis of the predicted results of climate conditions and salt migration behavior and soil parameters, formulates irrigation and soil conditioning strategies, and generates irrigation instructions, including adjusting the irrigation amount, irrigation time, irrigation method, and application of conditioning agents, so as to achieve precise improvement of saline-alkali land and ensure the optimal state of salt distribution and soil structure;

[0136] The irrigation execution module controls the operation of irrigation equipment according to the generated irrigation instructions, including the adjustment of water volume, irrigation time and irrigation method, to ensure the efficient use of water resources, and combines the prediction results to prevent excessive accumulation or loss of salt in the soil;

[0137] The data feedback module provides feedback on the irrigation execution results and the actual soil improvement effects. By real-time monitoring of irrigation and conditioning effects, it continuously optimizes irrigation and soil conditioning strategies to ensure the stable operation of the system under different environmental conditions.

[0138] Through the above contents, the intelligent, precise and dynamic optimization of saline-alkali land improvement has been achieved. According to real-time data and prediction results, irrigation and soil conditioning strategies can be flexibly adjusted to improve soil quality, optimize water resource utilization, and ensure efficient operation under different environmental conditions.

[0139] The climate prediction module includes:

[0140] Data collection: Obtain historical and current climate data from historical databases and real-time weather stations, including historical and real-time temperature, humidity, rainfall, wind speed, and evaporation;

[0141] Data preprocessing: standardize and denoise the collected climate data;

[0142] Climate prediction model construction: Use multivariate linear regression model to predict climate parameters, expressed as:

[0143] T t+1 =α1T t +α2H t +α3R t +1;

[0144] Among them, T t+1 is the predicted temperature value for the next time step, T t is the temperature value of the current time step, H t is the current humidity value, R t is the current rainfall, α1, α2, α3 are regression coefficients, and 1 is the error term;

[0145] H t+1 =β1H t +β2T t +β3W t +2;

[0146] Among them, H t+1 is the predicted humidity value for the next time step, W t is the current wind speed value, β1, β2, β3 are regression coefficients, and 2 is the error term;

[0147] R t+1 =γ1Rt +γ2T t +γ3W t +3;

[0148] Among them, R t+1 is the predicted rainfall value for the next time step, γ1, γ2, γ3 are regression coefficients, and 3 is the error term;

[0149] E t+1 =δ1E t +δ2T t +δ3W t +4;

[0150] Among them, E t+1 is the predicted value of evaporation at the next time step, δ1, δ2, δ3 are regression coefficients, and 4 is the error term;

[0151] Model optimization: Determine the regression coefficients using the least squares method, and adjust the model parameters through cross-validation and residual analysis to improve prediction accuracy;

[0152] Climate Conditions Forecasting: Use the optimized multivariate linear regression model to predict climate conditions for future time periods and generate predicted values ​​for temperature, humidity, rainfall, and evaporation in future time steps;

[0153] Through the above content, the relationship between multiple meteorological factors is comprehensively considered to accurately predict the future temperature, humidity, rainfall and evaporation. The prediction results will be used for subsequent salt migration prediction and irrigation scheduling to ensure the accurate improvement of saline-alkali land and efficient use of water resources.

[0154] The least squares method is used to determine the coefficients of the regression model. The goal is to minimize the sum of squared errors between the predicted values ​​and the actual values, expressed as:

[0155]

[0156] Among them, y i is the actual value of the ith sample (e.g., temperature), x i is the characteristic vector of the i-th sample (such as temperature, humidity, rainfall), β is the regression coefficient vector to be calculated, and n is the number of samples;

[0157]

[0158] Where X is the feature matrix, containing the feature vectors x of all samples i , y is a vector of target variable values, containing the actual values ​​y of all samples i ;

[0159] Cross-validation uses K-fold cross-validation to evaluate the predictive performance of the model, expressed as:

[0160]

[0161] Among them, MSE (k) is the mean square error on the kth fold, where K is the number of cross-validation folds (e.g. 5 or 10);

[0162] Residual analysis is the difference between the actual and predicted values ​​and is used to check the quality of fit of the model and is expressed as:

[0163]

[0164] Among them, e i is the residual of the ith sample, y i is the actual value of the ith sample, is the predicted value of the i-th sample;

[0165]

[0166] Where RSS is the residual sum of squares.

[0167] The soil detection module includes:

[0168] Sensor deployment: Multiple soil sensors are deployed in different areas and depths of saline-alkali land to collect real-time data on salt concentration, pH value, moisture content and organic matter content. Soil sensors include conductivity sensors, pH sensors, soil moisture sensors and near-infrared spectrum sensors;

[0169] Data collection: Regularly collect data on soil salt concentration, pH value, moisture content, and organic matter content;

[0170] Data processing: preliminary processing of the collected soil parameters, including data cleaning, denoising and standardization;

[0171] Through the above content, the accuracy and consistency of the data are ensured, thus providing a reliable basis for subsequent analysis and decision-making, improving the scientificity and accuracy of saline-alkali land improvement, and helping to achieve timely adjustment and optimization of soil conditions.

[0172] Electric field applications include:

[0173] Electric field application: An electric field of predetermined strength and direction is applied through a series of electrodes in the saline-alkali land to form an electric potential difference. The direction θ is determined by the relative positions of the electrodes and is expressed as:

[0174]

[0175] Where V is the voltage between electrodes, and d is the distance between electrodes;

[0176] Salt migration control: Under the action of the electric field, the salt ions in the soil will migrate in an orderly manner along the direction of the electric field to form an electroosmotic flow. According to the monitored salt concentration changes, the salt migration speed v is calculated in real time, expressed as:

[0177] v = μE;

[0178] Where, μ is the electrical mobility of salt ions, and E is the electric field strength;

[0179] Dynamically adjust the electric field: According to real-time soil parameters (such as salt concentration and soil conductivity), the intensity E′ and direction θ′ of the electric field are dynamically adjusted through a feedback control algorithm, which is expressed as:

[0180] E′(t)=E+k1ΔS(t);

[0181] θ′(t)=θ+k2Δσ(t);

[0182] Among them, ΔS(t) is the change of salt concentration monitored in real time, Δσ(t) is the change of soil conductivity, k1 and k2 are adjustment coefficients, and t is time;

[0183] Path control and optimization: By controlling the direction of the electric field, a directional salt migration path is formed to avoid excessive accumulation of salt on the soil surface or in the root zone, and the direction of the electric field is continuously adjusted using the gradient descent method to ensure that the salt migrates along the expected path to the target area (such as drainage area or low-impact area);

[0184] The gradient descent method specifically includes:

[0185] Objective function definition: The objective is to minimize the accumulation of salt concentration in the crop root zone, expressed as:

[0186]

[0187] Among them, S i (θ) is the salt concentration at the ith soil location, θ is the direction parameter of the electric field, and n is the number of soil locations;

[0188] Gradient calculation: Find the gradient of the objective function J(θ), expressed as:

[0189]

[0190] Gradient descent algorithm: Through the gradient descent method, the direction of the electric field θ is dynamically adjusted, which is expressed as:

[0191]

[0192] Among them, η is the learning rate, which controls the adjustment step size;

[0193] Dynamically update the electric field direction: At each time step, the electric field direction θ is updated using the gradient descent method, and the electric field strength and direction are recalculated, expressed as:

[0194]

[0195] Through continuous iteration, the salt migration path is gradually optimized to minimize the salt concentration in the target area (such as the crop root zone).

[0196] Through the above contents, it is possible to flexibly respond to different soil conditions and minimize the accumulation of salt in sensitive areas. This not only improves the accuracy of saline-alkali land improvement, but also effectively avoids negative impacts on crop root zones, ensuring the long-term stability and sustainability of soil improvement.

[0197] Salt migration monitoring includes:

[0198] Salt concentration monitoring: Conductivity sensors are placed at different soil depths to collect soil conductivity data at each depth in real time, expressed as:

[0199] S(z,t)=a·σ(z,t)+b;

[0200] Where S(z,t) is the salt concentration at time t and depth z, σ(z,t) is the conductivity at the corresponding depth, and a and b are empirical constants;

[0201] Migration velocity calculation: Based on the change in salt concentration at different time points, the migration velocity of salt is calculated and expressed as:

[0202]

[0203] Where v(z,t) is the migration velocity, is the rate of change of salt concentration with time, Δz is the depth interval;

[0204] Salt distribution analysis: The Kriging interpolation method is used to spatially interpolate the salt concentration data at different depths to generate a distribution map of salt in the soil profile, which is expressed as:

[0205]

[0206] in, is the estimated salt concentration at depth z, S(z i ) is the known depth z i The salt concentration at i is the interpolation weight, calculated by the Kriging algorithm;

[0207] Through the above content, we focus on the real-time collection and accurate analysis of salt data, ensure the controllability of the salt migration process and the reliability of prediction, provide accurate basic data for the formulation of subsequent soil improvement strategies, and significantly improve the scientificity and effectiveness of saline-alkali land improvement.

[0208] The electrochemical salt migration prediction model adopts the time convolution network (TCN) model, which includes:

[0209] Input data preprocessing: Soil parameters and predicted climate conditions are used as input data and standardized to ensure that data with different characteristics are processed at the same scale, expressed as:

[0210]

[0211] Among them, X is the original input data (including climate conditions and soil parameters), μ is the mean of the input data, σ is the standard deviation of the input data, and X′ is the standardized input data;

[0212] Multi-channel input convolution: Process time series inputs from multiple data channels (such as temperature, humidity, rainfall, soil salt concentration, etc.) and extract the spatiotemporal dependencies of each feature, expressed as:

[0213]

[0214] in, is the output of the kth convolution kernel at time step t, The convolution weight of the kth convolution kernel for the cth channel, is the input data segment (length d) of the cth channel at time step t, and C is the number of channels of the input data;

[0215] Causal convolution: Ensure that the convolution operation only considers the data of the current time step and before, avoiding the leakage of future information, expressed as:

[0216]

[0217] Among them, y t is the convolution output at time step t, W i is the weight of the i-th convolution kernel, x t-i is the input data of time step ti;

[0218] Weighted climate condition convolution layer: The weight parameter is introduced in the convolution layer to emphasize the different effects of different climate conditions (such as temperature, humidity, rainfall) on salt migration, which is expressed as:

[0219]

[0220] Among them, yt is the output at time step t, α j is the weight coefficient corresponding to the jth climate condition, W j is the weight of the jth convolution kernel, x t-j is the input data of time step tj;

[0221] Residual connection and skip connection: Residual connection and skip connection are introduced to enhance the depth of the network and the ability to capture long-term dependencies, expressed as:

[0222] y t ′=y t +x t ;

[0223] Among them, y t ′ is the output after residual connection, y t is the output of the current convolutional layer, x t is the input of the current time step;

[0224] y t ″=y t ′+f(x t-k );

[0225] Among them, y t ″ is the output after skip connection, f(·) is the ReLU activation function;

[0226] Salt migration prediction: Based on the output of multi-layer convolution, the salt migration behavior and distribution trend in the future time step are predicted, expressed as:

[0227]

[0228] Among them, S t+1 is the predicted value of salt migration in the next time step, W m is the weight of the mth prediction layer, y t ″ is the output of the last convolution layer, and M is the total number of prediction layers;

[0229] Through the above content, combined with multi-channel input, causal convolution and weighted climate condition convolution layers, the spatiotemporal dependencies and the influence of climatic factors in the salt migration process are accurately captured. In addition, the introduction of residual connections and jump connections enhances the model's ability to handle long-term dependencies and ensures the accuracy of salt migration prediction. Through the weighted combination of prediction layers, the model can flexibly adapt to different soil and climate conditions, providing more accurate and dynamic decision-making support for saline-alkali land improvement, significantly improving the overall efficiency and reliability of the system.

[0230] Feedback correction includes:

[0231] Real-time data acquisition: Obtain real-time monitored salt concentration and migration velocity data and compare them with the output of the electrochemical salt migration prediction model;

[0232] Error calculation: Calculate the error between the predicted value and the actual monitored value, expressed as:

[0233] (t) = S actual (t)-S predicted (t);

[0234] Where, (t) is the error value at time t, S actual (t) is the actual monitored salt concentration, S predicted (t) is the salt concentration predicted by the model;

[0235] Parameter adjustment: According to the calculated error, the model parameters are adjusted through the gradient descent algorithm to make the prediction model more accurate, which is expressed as:

[0236]

[0237] Among them, θ new is the updated model parameter, θ old is the original model parameter, η is the learning rate, is the gradient of the error with respect to the model parameters;

[0238] Model correction: Apply the adjusted parameters to the electrochemical salt migration prediction model to update the model's prediction capabilities in real time to make it closer to the actual monitoring data;

[0239] Through the above content, it is ensured that the electrochemical salt migration prediction model can adapt to the actual changes in soil and climate conditions, provide more accurate prediction results, and provide data support for the adjustment of saline-alkali land improvement strategies.

[0240] Data analysis modules include:

[0241] Data collection and standardization: Collect and standardize data on climate conditions, salt migration prediction results, and soil parameters to ensure that each data dimension is processed on a unified scale;

[0242] Multidimensional feature fusion: A weighted fusion method is used to integrate the standardized climate conditions, soil parameters, and salt migration prediction results into a comprehensive index to evaluate the current soil conditions and future trends, expressed as:

[0243] F = α1·C + α2·S + α3·T;

[0244] Among them, F is the integrated index after fusion, C is the climate condition, S is the salt migration prediction result, T is the soil parameter, and α1, α2, and α3 are the corresponding weight coefficients;

[0245] Irrigation and soil conditioning strategy formulation: Based on the comprehensive index F, irrigation and soil conditioning strategies are formulated through optimization algorithms to determine the optimal irrigation amount, irrigation time, irrigation method and application amount of conditioners. The optimization goal is to maximize crop yield and soil health, which is expressed as:

[0246]

[0247] Among them, G is the objective function, y i is the contribution of the ith decision variable (such as irrigation amount, time, etc.) to the target, λ i is the weight coefficient, c i is the cost, ν i To adjust the parameters;

[0248] Irrigation instruction generation: Generate irrigation instructions based on the optimization results, including adjusting the irrigation amount Q, irrigation time T, irrigation method M and the application amount A of the conditioner, expressed as:

[0249] Q = β1·F + γ1;

[0250] U = β2·F + γ2;

[0251] A=β3·F+γ3;

[0252]

[0253] Among them, Q is the irrigation amount, U is the irrigation time, A is the amount of conditioner applied, M is the irrigation method, β1, β2, β3 are decision coefficients, γ1, γ2, γ3 are adjustment constants, and F1 and F2 are critical values;

[0254] Through the above, the use of water resources and conditioners can be optimized, ensuring the efficiency and sustainability of saline-alkali land improvement, which will help achieve optimal crop growth and soil health under complex environmental conditions.

[0255] The critical values ​​F1 and F2 are determined based on historical data, including:

[0256] Input data preparation: Calculate the comprehensive index F corresponding to each historical data point and obtain a set of all F values ​​{F1, F2, ..., F n};

[0257] K-means clustering: Select k = 3 (three irrigation methods: drip irrigation, sprinkler irrigation, and infiltration irrigation), and divide the F value into three clusters, expressed as:

[0258]

[0259] in, is the position of the jth cluster center at the t+1th iteration, is the set of all data points belonging to the jth cluster at the tth iteration, It is the update formula of cluster center, that is, each iteration updates the position of cluster center by calculating the average value of all F values ​​in the cluster;

[0260] Cluster classification: After multiple iterations, it converges to a stable state, at which point each F value will belong to a certain cluster. Let the three cluster centers be μ1, μ2, and μ3, and μ1<μ2<μ3.

[0261] Determine the thresholds F1 and F2: The thresholds F1 and F2 are determined by the following formula:

[0262]

[0263] Irrigation execution modules include:

[0264] Instruction reception: receiving the generated irrigation instructions, including adjusting the irrigation amount Q, irrigation time T and the application amount A of the conditioner;

[0265] Equipment parameter setting: Automatically adjust the operating parameters of irrigation equipment according to the generated irrigation instructions, including:

[0266] Water volume control: The water volume control of irrigation equipment is based on pump flow rate regulation or valve opening control, expressed as:

[0267] Q = P·t;

[0268] Among them, Q is the water volume set in the irrigation instruction, P is the flow rate (unit is volume / time), and t is the irrigation duration;

[0269] Time setting: The irrigation time is set to the start time of irrigation. start and end time U end , expressed as:

[0270] U end =U start +U;

[0271] Where, U is the duration of irrigation;

[0272] Irrigation mode selection: According to the M value in the irrigation instruction, adjust the working mode of the irrigation equipment, expressed as:

[0273]

[0274] Through the above content, it is possible to dynamically adapt to different soil and climate conditions and select the most suitable irrigation method, thereby achieving efficient use of resources and maximizing irrigation effects. Through automated and intelligent operations, the system's response speed and execution accuracy are improved, which helps to continuously optimize the improvement process of saline-alkali land in complex environments.

[0275] The data feedback module includes:

[0276] Real-time monitoring and data collection: The sensor network monitors the soil moisture content, salt concentration, pH value and residual concentration of conditioners during irrigation in real time, and collects relevant data, including the immediate impact of irrigation amount, irrigation time and irrigation method on the soil;

[0277] Error calculation and evaluation: The monitoring data is compared with the expected irrigation effect and soil improvement target, and the error is calculated and expressed as:

[0278] i (t) = S actual,i (t)-S predicted,i (t);

[0279] in, i (t) is the error of the i-th monitoring parameter at time t, S actual,i (t) is the actual monitored value, S predicted,i (t) is the value predicted in the data analysis module;

[0280] Kalman filter correction and adjustment: Apply the Kalman filter algorithm to correct the monitoring data, reduce the interference of noise on the feedback signal, and adjust the irrigation strategy and soil conditioning strategy based on the corrected monitoring data, which is expressed as:

[0281]

[0282] in, is the predicted value of the current state, K k is the Kalman gain, z k is the actual measurement value, H is the observation matrix, P k|k-1 is the prediction covariance, R is the measurement noise covariance;

[0283] Strategy optimization and update: Based on the adjusted strategy parameters, the irrigation and soil conditioning strategies are updated and new irrigation instructions are generated. The adjusted strategies can more accurately adapt to the current soil conditions and climate conditions, and optimize the irrigation effect and soil improvement efficiency.

[0284] Through the above, it is possible to quickly respond to changes in soil and climate conditions and continuously optimize irrigation strategies. This precise real-time feedback mechanism significantly improves the efficiency and effectiveness of saline-alkali land improvement and ensures stable operation and optimal performance of the system in complex environments.

[0285] The present invention covers any substitution, modification, equivalent method and scheme made on the essence and scope of the present invention. In order to make the public have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, but those skilled in the art can fully understand the present invention without the description of these details. In addition, in order to avoid unnecessary confusion about the essence of the present invention, well-known methods, processes, procedures, components and circuits are not described in detail.

[0286] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. Intelligent irrigation and soil conditioning system for saline-alkali land improvement, characterized in that: It includes climate prediction module, soil detection module, electrochemical salt migration prediction module, data analysis module, irrigation execution module and data feedback module, among which; The climate prediction module predicts future climate conditions, including rainfall, temperature, humidity and evaporation, based on real-time climate data and historical climate data; The soil detection module monitors the soil parameters of saline-alkali land in real time, including salt concentration, pH value, moisture content and organic matter content; The electrochemical salt migration prediction module predicts the migration behavior of salt in the soil based on the soil parameters of saline-alkali land and the predicted climate conditions, specifically including: Electric field application: Apply an electric field of predetermined strength and direction in the soil to promote the orderly migration of salt in the soil. According to the monitored soil parameters, the electric field strength and direction are dynamically adjusted to control the movement path of the salt. Salt migration monitoring: responsible for real-time monitoring of the concentration changes and migration speed of salt in the soil, and analyzing the distribution of salt at different depths; Migration prediction: Combined with the predicted climate conditions, the migration behavior and distribution trend of salt are predicted through the electrochemical salt migration prediction model; Feedback correction: According to the real-time monitoring data of salt migration monitoring, the electrochemical salt migration prediction model is corrected and the model parameters are adjusted; The data analysis module comprehensively analyzes the predicted results of climate conditions and salt migration behavior and soil parameters, formulates irrigation and soil conditioning strategies, and generates irrigation instructions, including adjusting irrigation amount, irrigation time, irrigation method, and application of conditioning agents; The irrigation execution module controls the operation of the irrigation equipment according to the generated irrigation instructions, including the adjustment of water volume, irrigation time and irrigation method; The data feedback module provides feedback on the irrigation execution results and the actual soil improvement effects, and continuously optimizes the irrigation and soil conditioning strategies by real-time monitoring of the irrigation and conditioning effects.

2. The intelligent irrigation and soil conditioning system for improving saline-alkali land according to claim 1, characterized in that: The climate prediction module includes: Data collection: Obtain historical and current climate data from historical databases and real-time weather stations, including historical and real-time temperature, humidity, rainfall, wind speed, and evaporation; Data preprocessing: standardize and denoise the collected climate data; Climate prediction model construction: Use multivariate linear regression model to predict climate parameters, expressed as: T t+1 =α1T t +α2H t +α3R t +1; Among them, T t+1 is the predicted temperature value for the next time step, T t is the temperature value of the current time step, H t is the current humidity value, R t is the current rainfall, α1, α2, α3 are regression coefficients, and 1 is the error term; H t+1 =β1H t +β2T t +β3W t +2; Among them, H t+1 is the predicted humidity value for the next time step, W t is the current wind speed value, β1, β2, β3 are regression coefficients, and 2 is the error term; R t+1 =γ1R t +γ2T t +γ3W t +3; Among them, R t+1 is the predicted rainfall value for the next time step, γ1, γ2, γ3 are regression coefficients, and 3 is the error term; E t+1 =δ1E t +δ2T t +δ3W t +4; Among them, E t+1 is the predicted value of evaporation at the next time step, δ1, δ2, δ3 are regression coefficients, and 4 is the error term; Model optimization: Determine regression coefficients using the least squares method, and adjust model parameters through cross-validation and residual analysis; Climate Conditions Prediction: Use an optimized multivariate linear regression model to predict climate conditions for future time periods and generate predicted values ​​of temperature, humidity, precipitation, and evaporation for future time steps.

3. The intelligent irrigation and soil conditioning system for improving saline-alkali land according to claim 1, characterized in that: The soil detection module comprises: Sensor deployment: Multiple soil sensors are deployed in different areas and depths of saline-alkali land to collect real-time data on salt concentration, pH value, moisture content and organic matter content. Soil sensors include conductivity sensors, pH sensors, soil moisture sensors and near-infrared spectrum sensors; Data collection: Regularly collect data on soil salt concentration, pH value, moisture content, and organic matter content; Data processing: preliminary processing of the collected soil parameters, including data cleaning, denoising and standardization.

4. The intelligent irrigation and soil conditioning system for improving saline-alkali land according to claim 1, characterized in that: The electric field application includes: Electric field application: An electric field of predetermined strength and direction is applied through a series of electrodes in the saline-alkali land to form an electric potential difference. The direction θ is determined by the relative positions of the electrodes and is expressed as: Where V is the voltage between electrodes, and d is the distance between electrodes; Salt migration control: Under the action of the electric field, the salt ions in the soil will migrate in an orderly manner along the direction of the electric field to form an electroosmotic flow. According to the monitored salt concentration changes, the salt migration speed v is calculated in real time, expressed as: v = μE; Where, μ is the electrical mobility of salt ions, and E is the electric field strength; Dynamically adjust the electric field: According to the real-time soil parameters, the intensity E′ and direction θ′ of the electric field are dynamically adjusted through the feedback control algorithm, which is expressed as: E′(t)=E+k1ΔS(t); θ′(t)=θ+k2Δσ(t); Among them, ΔS(t) is the change of salt concentration monitored in real time, Δσ(t) is the change of soil conductivity, k1 and k2 are adjustment coefficients, and t is time; Path control and optimization: By controlling the direction of the electric field, a directional salt migration path is formed, and the direction of the electric field is continuously adjusted using the gradient descent method.

5. The intelligent irrigation and soil conditioning system for improving saline-alkali land according to claim 4, characterized in that: The salt migration monitoring includes: Salt concentration monitoring: Conductivity sensors are placed at different soil depths to collect soil conductivity data at each depth in real time, expressed as: S(z,t)=a·σ(z,t)+b; Where S(z,t) is the salt concentration at time t and depth z, σ(z,t) is the conductivity at the corresponding depth, and a and b are empirical constants; Migration velocity calculation: Based on the change in salt concentration at different time points, the migration velocity of salt is calculated and expressed as: Where v(z,t) is the migration velocity, is the rate of change of salt concentration with time, Δz is the depth interval; Salt distribution analysis: The Kriging interpolation method is used to spatially interpolate the salt concentration data at different depths to generate a distribution map of salt in the soil profile, which is expressed as: in, is the estimated salt concentration at depth z, S(z i ) is the known depth z i The salt concentration at i is the interpolation weight.

6. The intelligent irrigation and soil conditioning system for improving saline-alkali land according to claim 5, characterized in that: The electrochemical salt migration prediction model adopts a time convolutional network model, and the time convolutional network model includes: Input data preprocessing: soil parameters and predicted climate conditions are used as input data and standardized; Multi-channel input convolution: Process time series input from multiple data channels and extract the spatiotemporal dependencies of each feature, expressed as: in, is the output of the kth convolution kernel at time step t, The convolution weight of the kth convolution kernel for the cth channel, is the input data fragment of the cth channel at time step t, and C is the number of channels of the input data; Causal convolution: Ensure that the convolution operation only considers the data of the current time step and before, expressed as: Among them, y t is the convolution output at time step t, W i is the weight of the i-th convolution kernel, x t-i is the input data of time step ti; Weighted climate condition convolution layer: A weight parameter is introduced in the convolution layer to emphasize the different effects of different climate conditions on salt migration, expressed as: Among them, y t is the output at time step t, α j is the weight coefficient corresponding to the jth climate condition, W j is the weight of the jth convolution kernel, x t-j is the input data of time step tj; Residual connection and skip connection: Residual connection and skip connection are introduced to enhance the depth of the network and the ability to capture long-term dependencies, expressed as: and t 'and t +x t ; Among them, y t ′ is the output after residual connection, y t is the output of the current convolutional layer, x t is the input of the current time step; y t ″ =y t ′+f(x t-k ); Among them, y t ″ is the output after skip connection, f(·) is the ReLU activation function; Salt migration prediction: Based on the output of multi-layer convolution, the salt migration behavior and distribution trend in the future time step are predicted, expressed as: Among them, S t+1 is the predicted value of salt migration in the next time step, W m is the weight of the mth prediction layer, y t ″ is the output of the last convolution layer, and M is the total number of prediction layers.

7. The intelligent irrigation and soil conditioning system for improving saline-alkali land according to claim 6, characterized in that: The feedback correction includes: Real-time data acquisition: Obtain real-time monitored salt concentration and migration velocity data and compare them with the output of the electrochemical salt migration prediction model; Error calculation: Calculate the error between the predicted value and the actual monitored value, expressed as: (t)=S actual (t)-S predicted (t); Where, (t) is the error value at time t, S actual (t) is the actual monitored salt concentration, S predicted (t) is the salt concentration predicted by the model; Parameter adjustment: According to the calculated error, the model parameters are adjusted through the gradient descent algorithm, which is expressed as: Among them, θ new is the updated model parameter, θ old is the original model parameter, η is the learning rate, is the gradient of the error with respect to the model parameters; Model calibration: Apply the adjusted parameters to the electrochemical salt migration prediction model to update the model's predictive capabilities in real time.

8. The intelligent irrigation and soil conditioning system for improving saline-alkali land according to claim 1, characterized in that: The data analysis module includes: Data collection and standardization: Collect and standardize data on climate conditions, salt migration predictions, and soil parameters; Multidimensional feature fusion: A weighted fusion method is used to integrate the standardized climate conditions, soil parameters, and salt migration prediction results into a comprehensive index to evaluate the current soil conditions and future trends, expressed as: F = α1·C + α2·S + α3·T; Among them, F is the integrated index after fusion, C is the climate condition, S is the salt migration prediction result, T is the soil parameter, and α1, α2, and α3 are the corresponding weight coefficients; Irrigation and soil conditioning strategy formulation: Based on the comprehensive index F, irrigation and soil conditioning strategies are formulated through optimization algorithms to determine the optimal irrigation amount, irrigation time, irrigation method and application amount of conditioners. The optimization goal is to maximize crop yield and soil health, which is expressed as: Among them, G is the objective function, y i is the contribution of the i-th decision variable to the target, λ i is the weight coefficient, c i is the cost, ν i To adjust the parameters; Irrigation instruction generation: Generate irrigation instructions based on the optimization results, including adjusting the irrigation amount Q, irrigation time T, irrigation method M and the application amount A of the conditioner, expressed as: Q = β1·F + γ1; U = β2·F + γ2; A=β3·F+γ3; Among them, Q is the irrigation amount, U is the irrigation time, A is the amount of conditioner applied, M is the irrigation method, β1, β2, β3 are decision coefficients, γ1, γ2, γ3 are adjustment constants, and F1 and F2 are critical values.

9. The intelligent irrigation and soil conditioning system for improving saline-alkali land according to claim 8, characterized in that: The irrigation execution module comprises: Instruction reception: receiving the generated irrigation instructions, including adjusting the irrigation amount Q, irrigation time T and the application amount A of the conditioner; Equipment parameter setting: Automatically adjust the operating parameters of irrigation equipment according to the generated irrigation instructions, including: Water volume control: The water volume control of irrigation equipment is based on pump flow rate regulation or valve opening control, expressed as: Q = P·t; Among them, Q is the water volume set in the irrigation instruction, P is the flow rate, and t is the irrigation duration; Time setting: The irrigation time is set to the start time of irrigation. start and end time U end , expressed as: IN end =U start +U; Where, U is the duration of irrigation; Irrigation mode selection: According to the M value in the irrigation instruction, adjust the working mode of the irrigation equipment, expressed as:

10. The intelligent irrigation and soil conditioning system for improving saline-alkali land according to claim 1, characterized in that: The data feedback module comprises: Real-time monitoring and data collection: The sensor network monitors the soil moisture content, salt concentration, pH value and residual concentration of conditioners during irrigation in real time, and collects relevant data, including the immediate impact of irrigation amount, irrigation time and irrigation method on the soil; Error calculation and evaluation: The monitoring data is compared with the expected irrigation effect and soil improvement target, and the error is calculated and expressed as: i (t)=S actual,i (t)-S predicted,i (t); in, i (t) is the error of the i-th monitoring parameter at time t, S actual,i (t) is the actual monitored value, S predicted,i (t) is the value predicted in the data analysis module; Kalman filter correction and adjustment: Apply the Kalman filter algorithm to correct the monitoring data, and adjust the irrigation strategy and soil conditioning strategy based on the corrected monitoring data; Strategy optimization and update: Based on the adjusted strategy parameters, the irrigation and soil conditioning strategies are updated and new irrigation instructions are generated.

Images