Daily scale optimal irrigation system determination method based on Jensen model-dynamic programming-crop growth model coupling

Abstract

The invention discloses a day scale optimal irrigation system determination method based on Jensen model-dynamic programming-crop growth model coupling. The method comprises the following steps: acquiring research area data; carrying out data processing to obtain day-by-day input data of the dynamic planning model, wherein the day-by-day input data comprises the daily effective rainfall, the daily planned wetting layer depth, the daily effective rainfall of the planned wetting layer, the daily moisture sensitivity index and the daily maximum evapotranspiration of crops; developing an irrigation system optimization program for daily calculation based on a Jensen model and a dynamic programming algorithm, and calculating a daily scale optimal irrigation scheme based on the Jensen model under a water limiting condition; adjusting the irrigation scheme according to the single minimum irrigation amount to obtain a better solution set; inputting the better solution set into the verified crop growth model to calculate the crop yield and the water utilization efficiency to obtain a Pareto optimal solution set; and an optimal daily scale irrigation system is selected. The invention provides a method which gives consideration to irrigation system optimization and growth condition simulation and can provide a daily-scale irrigation system for any crop.

Description

TECHNICAL FIELD

[0001] The application relates to a method for determining a daily scale optimal irrigation system based on coupling of a Jensen model-dynamic programming-crop growth model, and belongs to the field of agricultural intelligent irrigation optimization and decision-making. BACKGROUND

[0002] Traditional irrigation system optimization methods (dynamic programming DP, genetic algorithm, etc.) are mostly based on field test data, adopt suitable water production functions, combine linear and nonlinear regression, least square method, etc. to calculate the water sensitive index, and then calculate the optimal allocation of irrigation water. The calculation result is the irrigation water amount of each growth period of crops, and specific irrigation dates cannot be determined. Crop growth models such as AquaCrop are water-driven crop growth models, the core mechanism of which is to calculate the accumulation of biomass through transpiration, and not based on traditional water production functions (Blank, Jensen model, etc.). The advantages of such models are that the model can be calibrated and verified based on field test data, so as to simulate the crop growth results of a specific irrigation scheme. The disadvantage is that the irrigation system cannot be optimized; or the crop yield and water use efficiency can be calculated for each input irrigation scheme, and the optimal one is selected, but the trial calculation amount is huge; in addition, the scientificity and rationality of the input irrigation scheme of the model need to be studied. At present, crop growth models are mostly used in scientific research, and are not easy for farmers to master, which is limited in localization, applicability and popularization.

[0003] Dynamic programming method is widely used in irrigation system optimization field due to its easy-to-construct conceptual model and solution, but it is mostly based on the calculation of suitable irrigation amount in the growth period of crops, and less in daily scale, and cannot determine the specific irrigation day. In addition, the dynamic programming method cannot simulate the results of field test scheme, which is not as good as the crop growth model.

[0004] Therefore, in order to solve the above problems, the crop water production function, dynamic programming and crop growth model are coupled, an optimal irrigation system in daily scale which is easy for farmers to understand is proposed, the dual needs of irrigation system optimization and growth simulation are met, the short boards are complemented, and the respective advantages are exerted. SUMMARY

[0005] To solve the above technical problems, the application provides a method for determining a daily scale optimal irrigation system based on coupling of a Jensen model-dynamic programming-crop growth model, so as to make up for the shortcomings of the existing irrigation system optimization method.

[0006] To achieve the above object, the application provides a daily scale optimal irrigation system determination method based on coupling of Jensen model-dynamic programming-crop growth model, comprising the following steps:

[0007] Step 1: obtaining meteorological, soil, crop parameter, field test and other data of a research area;

[0008] Step 2: processing the data to obtain daily input data of the dynamic programming model: daily effective rainfall, daily planned wet layer depth, daily effective rainfall of the planned wet layer, daily water sensitive index of the crop, daily maximum evapotranspiration of the crop;

[0009] Step 3: developing a daily calculation irrigation system optimization program based on Jensen model and dynamic programming algorithm, and calculating a daily scale optimal irrigation scheme based on Jensen model under water limiting conditions;

[0010] Step 4: further adjusting the irrigation scheme according to the single minimum irrigation amount constraint to obtain a better solution set;

[0011] Step 5: calibrating and verifying the crop growth model according to the test data;

[0012] Step 6: inputting the better solution set into the calibrated and verified crop growth model to calculate the crop yield and water use efficiency, and obtaining a Pareto optimal solution set;

[0013] Step 7: selecting and determining the optimal irrigation system from the Pareto optimal solution set in combination with the entropy weight method.

[0014] The step 1 comprises obtaining crop yield, actual water consumption, crop daily planned wet layer depth (which can be valued according to local experience), soil bulk density, field moisture capacity, wilting coefficient, suitable soil moisture content range of the crop and other related data of a test station in a research year; and obtaining meteorological data such as daily temperature, rainfall, wind speed and sunshine duration of the research area in a research year.

[0015] The step 2 comprises:

[0016] Step 201: calculating the daily effective rainfall in the planned wet layer.

[0017] Firstly, the soil effective water storage capacity V of the planned wet layer is calculated i :

[0018]

[0019] In the formula, γ is the dry bulk density of soil, g / cm 3 ; H i is the planned wet layer depth on the i-th day, m; and θ is the field moisture capacity.a The lower limit of soil moisture content is 65% of the upper limit (field capacity).

[0020] The planned wet layer depth is determined by experience, measured data or formula calculation.

[0021] Secondly, the daily effective rainfall (excluding runoff, ineffective evaporation and crop interception) is calculated according to the effective rainfall coefficient;

[0022] Finally, the daily effective rainfall in the planned wet layer is calculated. The specific method is to compare the daily effective rainfall with the effective water storage capacity of the planned wet layer.

[0023] If the daily effective rainfall is greater than the effective water storage capacity of the planned wet layer, the daily effective rainfall of the planned wet layer is equal to the effective water storage capacity of the planned wet layer, and the excess water seeps into the lower layer.

[0024] On the contrary, if the daily effective rainfall is less than or equal to the effective water storage capacity of the planned wet layer, the daily effective rainfall of the planned wet layer is equal to the daily effective rainfall.

[0025] Assumptions: (1) The daily effective rainfall of the planned wet layer can be completely used for evapotranspiration, even if it cannot be completely used for evapotranspiration in the day, it will be converted into evapotranspiration in the following days. (2) The soil moisture content is not less than 65% of the field capacity.

[0026] Step 202: Calculate the daily water sensitive index of the crop. The change process of the cumulative value of the water sensitive index with time is represented by the Logisitic growth curve:

[0027]

[0028] In the formula: is the cumulative value of the water sensitive index; t is the growth days of the crop from the sowing day, and its change process is from 0 to the length of the crop growth period T; 、 、 is a to-be-determined coefficient.

[0029] According to the yield, water consumption and maximum yield under sufficient water supply conditions measured by the test station, the least square principle is applied to adjust the parameters 、 、 , and the Logisitic growth curve is fitted so that the error sum of squares of the relative yield calculated by the Jensen model and the actual relative yield is minimized. The daily water sensitive index is equal to the difference between the cumulative functions at the beginning and end of the day.

[0030] The actual value of the relative yield of all treatments ) and relative output calculation ( The sum of squared errors:

[0031]

[0032] In the formula: The number of items to be processed; This refers to the actual output of each process; The yield for each treatment is calculated; This represents the maximum output in practice.

[0033] Step 203: Calculate the potential evapotranspiration ET using the Penman-Montes formula. 0i The maximum daily evapotranspiration ET of crops is calculated by combining the crop coefficient. mi .

[0034] Step 3 involves constructing a daily-scale irrigation regime optimization model based on the Jensen model and dynamic programming algorithm, including:

[0035] Step 301: Determine the state variables. State variables include the allocatable water quantity q at the beginning of each stage (initial of each day). i And the amount of water available to crops within the planned wetting layer W i , i = 1, 2, ..., N, W i It is a function of soil moisture content.

[0036]

[0037] In the formula: γ is the dry bulk density of the soil, g / cm³ 3 H i Let m be the planned wetting layer depth for day i.

[0038] θ i Let be the average soil moisture content within the planned wetting layer on day i, expressed as a percentage of dry weight, % .

[0039] θ a The lower limit of soil moisture content is %, and the percentage is 65% of the upper limit (field water holding capacity).

[0040] Step 302: Determine the decision variables. There are two decision variables: the actual daily irrigation volume (m). i and actual evaporation ET ai , i = 1, 2, …, N.

[0041] Step 303: Establish the state transition equations. The state transition equations are the water distribution equation and the water balance equation within the planned wetting layer, describing the relationship between the variables during the state transition process.

[0042] Water distribution equation:

[0043]

[0044] The water balance equation within the planned wetting layer is as follows:

[0045]

[0046] In the formula: This represents the actual daily water consumption of the crop, in mm. = K si ·K ci ·ET0, where K ci K represents the crop coefficient for day i, using the recommended value from FAO Irrigation and Drainage Paper No. 56 or the calculated value from the experimental station; si K is the water stress coefficient; when there is no stress, K si =1; P i Daily effective rainfall, mm; C i The daily leakage exceeding the planned amount in the wetting layer, in mm; m i Let q be the irrigation amount on day i, in mm; i q i＋1 The amount of water available for irrigation at the beginning of day i and day i+1, in mm; The field water use coefficient varies significantly depending on the irrigation method. The value should be determined based on the actual situation.

[0047] Step 304: Determine the constraints.

[0048] Irrigation water volume and evapotranspiration constraints:

[0049]

[0050]

[0051]

[0052]

[0053] In the formula: Q is the total amount of water available for irrigation, in mm; Minimum daily evapotranspiration, in mm; This represents the maximum evapotranspiration on day i, in mm. Other values ​​have the same meaning as above.

[0054] Soil moisture content constraints:

[0055]

[0056] In the formula: This represents the lower limit of soil moisture content, expressed as a percentage, and is taken as 65% of the upper limit (field holding capacity). This refers to the field water holding capacity.

[0057] Step 305: Establish the objective function.

[0058] Given a limited supply of water, the objective is to maximize the relative yield of the Jensen model, i.e., Y. a / Y m It tends to the maximum value of 1.

[0059]

[0060] In the formula, N is the total number of days for crop growth; i is the i-th day of crop growth; ET ai ET represents the actual evapotranspiration on day i, in mm. mi Y represents the maximum evapotranspiration on day i, in mm. a The actual yield of the crop, kg / hm 2 ;Y m The maximum yield of a crop, i.e., the crop yield under conditions of sufficient water supply, is expressed in kg / hm². 2 ;λ i Let be the moisture sensitivity index on day i.

[0061] When irrigation is fully irrigated, the theoretical value of the Jensen model equals 1; when irrigation is not fully irrigated, the goal is to maximize the Jensen function value, i.e., the optimal irrigation regime, which is a multi-stage decision optimization process. Thus, the optimization process of the irrigation regime is transformed into the solution process of the Jensen model.

[0062] Step 306: Solution Process. A successive approximation method with dimensionality reduction is used to solve the problem. First, the daily soil moisture content is initially estimated. The available water supply is discretized, and the actual evapotranspiration corresponding to different irrigation volumes is calculated in reverse. The initial irrigation allocation is determined using the cumulative benefit maximization principle of the Jensen model. Then, based on this allocation, the dynamic changes in soil moisture content are calculated in forward direction. Evapotranspiration is adjusted according to the moisture content constraint. The calculated soil moisture content is compared with the initial value, the initial value is modified, and the calculation is repeated until the absolute error between the calculated and initial estimates of the moisture content is less than 0.0001. A dynamic programming program is developed using Matlab or Visual Studio software to solve the above model.

[0063] Step 4 includes determining the minimum irrigation volume per application based on different irrigation methods (furrow irrigation, sprinkler irrigation, drip irrigation). Adjust the daily irrigation volume calculation results. If the calculated value for a certain day... If the amount of water used is less than the minimum irrigation amount for a certain irrigation method, then the irrigation amounts for day i and the next 5 days are added together to obtain the result. ,like Then the irrigation amount on day i is Otherwise, the irrigation amount on day i is set to 0. This yields a better solution set.

[0064] Then, based on steps 5, 6, and 7, the optimal irrigation system is obtained.

[0065] Compared with the prior art, the present invention has the following advantages and technical effects:

[0066] This invention provides a method for determining the optimal irrigation regime at the daily scale based on the coupling of the Jensen model, dynamic programming, and a crop growth model. By coupling the crop water production function, dynamic programming, and the crop growth model, the optimal irrigation regime at the daily scale can be calculated. This method optimizes the irrigation regime based on the crop water production function and uses dynamic programming algorithms to calculate at the daily scale, making it relatively easy for growers to understand. The crop growth model can simulate the actual growth of crops, facilitating the mutual verification of theory and practice. The coupling of the crop water production function, dynamic programming, and crop growth model satisfies the dual needs of optimizing the irrigation regime and simulating crop growth, complementing each other's weaknesses and leveraging their respective advantages. Theoretically, this method can provide a daily-scale optimized irrigation regime scheme for any crop, enriching the methods for optimizing high-yield and high-efficiency irrigation regimes under water-limited conditions, and has important practical reference value for the widespread implementation of smart irrigation. Attached Figure Description

[0067] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0068] Figure 1 This is a flowchart of a method for determining the optimal irrigation regime on a daily scale based on the coupling of Jensen model, dynamic programming, and crop growth model according to the present invention.

[0069] Figure 2 Here is a flowchart of a multidimensional dynamic programming algorithm based on successive approximation;

[0070] Figure 3 This is a graph showing the daily water sensitivity index λ and the cumulative water sensitivity index Z(t) of summer maize in an embodiment of the present invention.

[0071] Figure 4This invention presents a daily-scale optimal irrigation scheme based on the Jensen model, as an embodiment of the present invention.

[0072] Figure 5 This is the actual evapotranspiration under the optimal daily-scale irrigation scheme based on the Jensen model in this embodiment of the invention.

[0073] Figure 6 This embodiment of the invention provides soil moisture content based on the optimal daily-scale irrigation scheme using the Jensen model.

[0074] Figure 7 This is a Pareto front for multi-objective optimization of daily-scale irrigation regimes based on the Jensen-DP-AquaCrop model coupling, as described in this embodiment of the invention. Detailed Implementation

[0075] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments:

[0076] The optimal irrigation regime at the daily scale is calculated based on a method for determining the optimal irrigation regime at the daily scale that is based on the coupling of the Jensen model, dynamic programming, and crop growth model.

[0077] Step 1: Obtain meteorological, soil, crop parameters, and field trial data for the study area;

[0078] We collected and compiled measured and experimental data on summer maize in 2024 from the Baoding Irrigation Experimental Station in Hebei Province. Crop yield, actual water consumption, soil bulk density, and field water holding capacity were all obtained from field measurements. Daily planned wetting depth and suitable soil moisture content range were determined based on literature review and the station's actual conditions or experience. Meteorological data were obtained from meteorological station observations. In this case, furrow irrigation was used, with a water supply of 60 mm. The average precipitation during the summer maize growing season in 2024 was 710.2 mm, corresponding to a frequency close to P=2%, indicating an extremely wet year. The growth stages were: seedling stage - jointing (June 19 - July 24), jointing stage - tasseling stage (July 25 - August 24), tasseling stage - grain filling stage (August 25 - September 2), and grain filling stage - maturity stage (September 3 - October 2).

[0079] Step 2: Process the data to obtain the daily input data for the dynamic programming model: daily effective rainfall, daily planned wetting layer depth, daily effective rainfall of the planned wetting layer, daily water sensitivity index of crops, and daily maximum evapotranspiration of crops;

[0080] Step 201: Calculate the effective soil water storage capacity V of the planned wetting layer based on the data from the experimental station. iThe daily effective rainfall is calculated based on the daily rainfall and effective rainfall coefficient in the meteorological data. The effective rainfall coefficient is set as follows: 0 for daily rainfall of 0-3 mm; 0.85 for 3-50 mm; and 0.7 for >50 mm. Furthermore, the daily effective rainfall is compared with the effective soil water storage capacity of the planned wetting layer to calculate the daily effective precipitation within the planned wetting layer.

[0081] Step 202: Based on the yield, water consumption, and maximum yield under sufficient water supply conditions for each treatment at the experimental station, apply the least squares principle to fit the Logisitic growth curve, minimizing the sum of squared errors between the relative yield calculated by the Jensen model and the actual relative yield. Further based on... Calculate the daily water sensitivity index of crops .

[0082] Step 203: Calculate the potential evapotranspiration ET using the Penman-Montes formula. 0i The crop coefficient K was calculated by combining data from the experimental station. ci Further calculate the maximum daily evapotranspiration of crops ET mi = K ci ET 0i .

[0083] Step 3: Develop a daily-scale irrigation regime optimization program based on the Jensen model and dynamic programming algorithm to calculate the optimal irrigation scheme on a daily scale under water-limited conditions;

[0084] Step 301: Determine the state variables. State variables include the initial daily allocable water quantity q. i And the amount of water available to crops within the planned wetting layer W i Since the first part is a reverse calculation, the initial moisture content is set to 65% of the field capacity. The amount of water W available to the crop within the planned wetting layer on the last day is then calculated. i On the last day, the initial amount of water available for allocation is equal to the amount of water available for supply.

[0085] Step 302: Determine the decision variables. There are two decision variables: the actual daily irrigation volume (m). i and actual evaporation ET ai , i = 1, 2, …, N.

[0086] Step 303: Establish the state transition equation. Calculate the daily values ​​based on the water balance equation within the planned wetting layer.

[0087] Step 304: Write out the constraints based on the known data.

[0088] Step 305: Establish the objective function.

[0089] Step 306: Solution Process. A successive approximation method with dimensionality reduction is used to solve the problem. First, the daily soil moisture content is initially estimated. The available water supply is discretized, and the actual evapotranspiration corresponding to different irrigation volumes is calculated in reverse. The initial irrigation allocation is determined using the cumulative benefit maximization principle of the Jensen model. Then, based on this allocation, the dynamic changes in soil moisture content are calculated in forward direction. Evapotranspiration is adjusted according to the moisture content constraint. The calculated soil moisture content is compared with the initial value, the initial value is modified, and the calculation is repeated until the absolute error between the calculated and initial estimates of the moisture content is less than 0.0001. A dynamic programming program is developed using Matlab or Visual Studio software to solve the above model.

[0090] The dynamic programming calculation program defines 8 constants (number of growing days N, total irrigation Q, number of discrete irrigation times NT1, field water holding capacity FWC, lower limit of soil moisture content WWC, soil dry bulk density SOIL, irrigation coefficient FACTOR) and 4 variables (water sensitivity index [N], planned wetting layer depth h [N+1], effective rainfall within the planned wetting layer p [N], effective rainfall pz [N], maximum evapotranspiration emax [N], minimum evapotranspiration emin [N]).

[0091] In 2024, the growing season for summer maize at the Baoding Irrigation Experimental Station was 106 days, with a total irrigation volume of 100 mm (i.e., 1000 m³). 3 · hm -2 ), using 10mm as the unit (can be set to any positive value), the discrete irrigation amount, then the discrete number NT1=10; the soil dry bulk density is 1.58g / cm³. 3 The irrigation water utilization coefficient for furrow irrigation is taken as 0.85. That is: N=106, Q=100, NT1=10; FACTOR=0.95, SOIL=1.58; FWC=0.221, WWC=0.144 (WWC is 65% of FWC); the daily emin[N] is taken as 0.05; the remaining data are stored in the form of an Excel spreadsheet, including: index[N], h[N+1], p[N], pz[N], emax[N].

[0092] First, the first part is a reverse calculation, starting from day N. The estimated daily soil moisture content (all initial estimates are 65% of field capacity) and irrigation amount gd = i·Q / NT1 are substituted into the state transition equation to calculate the actual daily evapotranspiration. After multiple calculations, a preliminary irrigation allocation is determined based on the principle of maximizing the cumulative benefit of the Jensen model in step 305. Second, the second part is a forward calculation, which calculates the dynamic changes in soil moisture content based on the irrigation allocation values ​​from the first part. The evapotranspiration value is adjusted according to the moisture content constraint. The calculated soil moisture content value is compared with the initial value, the initial value is modified, and the calculation is repeated. The calculation stops when the absolute error between the calculated and initial estimates of the moisture content is less than 0.0001, and the result is output, thus obtaining the daily-scale irrigation regime.

[0093] Part of the code is as follows:

[0094] % Total water inflow into the soil that day

[0095] total_water_input = fmax1(k, 2) * FACTOR + p(k) + B(k);

[0096] % --- Treatment of soil moisture content below the wilting point ---

[0097] if swc_next_theoretical < WWC

[0098] % Scenario 1: Severe water deficit, soil water level drops to WWC

[0099] swc(k+1) = WWC;

[0100] % Calculate the maximum possible water consumption for the day (from the current state down to WWC).

[0101] max_water_extraction = 1000 * h(k+1) * SOIL * (swc(k)- WWC);

[0102] The actual evaporation rate must not exceed the sum of the maximum possible consumption and the total input water volume.

[0103] Furthermore, due to water shortages, the actual evapotranspiration will be less than the demand (emax), and may even be less than emin.

[0104] actual_evap = min(emax(k), max_water_extraction +total_water_input);

[0105] actual_evap = max(emin(k), actual_evap); % Ensures that the actual_evap value is not lower than the minimum evaporation rate.

[0106] In this situation, no drainage occurs, and all the water is consumed.

[0107] drainsp(k) = 0;

[0108] elseif swc_next_theoretical > FWC

[0109] % Situation 2: Excessive moisture, requiring drainage

[0110] swc(k+1) = FWC;

[0111] actual_evap = emax(k); % With sufficient moisture, evaporate at maximum capacity.

[0112] % Calculate drainage volume (positive value)

[0113] drainsp(k) = 1000 * h(k+1) * SOIL * (swc(k) - FWC) +...

[0114] total_water_input - emax(k);

[0115] else

[0116] % Situation 3: Normal moisture content, no stress, no drainage

[0117] swc(k+1) = swc_next_theoretical;

[0118] actual_evap = 1000 * h(k+1) * SOIL * (swc(k) - swc(k+1)) + ...

[0119] total_water_input;

[0120] Theoretically, actual_evap should equal emax(k), but the calculation may have slight errors and needs to be constrained.

[0121] if actual_evap > emax(k)

[0122] actual_evap = emax(k);

[0123] elseif actual_evap < emin(k)

[0124] actual_evap = emin(k);

[0125] end

[0126] Within the normal range, the drainage volume should be 0 or a very small value (calculation error).

[0127] drainsp(k) = max(0, 1000 * h(k+1) * SOIL * (swc(k) -swc(k+1)) + total_water_input - actual_evap);

[0128] end

[0129] % Calculate the daily benefits

[0130] daily_benefit = (actual_evap / emax(k)) ^ index(k);

[0131] % Calculate total benefit

[0132] total_benefit = total_benefit * daily_benefit;

[0133] fmax2(k, 1) = total_benefit; % Cumulative benefit

[0134] fmax2(k, 2) = actual_evap; % Actual evaporation rate

[0135] Those skilled in the art will understand that the above code is merely an exemplary implementation, and the same function can be achieved using other programming languages ​​or logical structures without departing from the core idea of ​​this invention.

[0136] Step 4: Further adjust the irrigation scheme based on the minimum irrigation volume constraint to obtain a better solution set; based on the minimum irrigation volume of different irrigation methods ( Adjust the daily irrigation volume calculation results. If the calculated value for a certain day... If the amount of water used is less than the minimum irrigation amount for a certain irrigation method, then the irrigation amounts for day i and the next 5 days are added together to obtain the result. ,like Then the irrigation amount on day i is Otherwise, the irrigation amount on day i is set to 0. This yields a better solution set.

[0137] Table 1

[0138]

[0139] Step 5: Calibrate and validate the crop growth model based on experimental data; calibrate the AquaCrop model using measured data from the experimental station in 2023 and validate it using data from 2024.

[0140] Step 6: Input the optimal solution set into the calibrated and verified crop growth model to calculate crop yield and water use efficiency, and obtain the Pareto optimal solution set; input the irrigation scheme of the optimal solution set obtained in Step 4 into the calibrated and verified AquaCrop model in Step 5 to calculate crop yield and water use efficiency, and analyze to obtain the Pareto optimal solution set; the objective function can be single-objective or multi-objective; multi-objectives are not limited to yield and water use efficiency, and other objectives such as economic benefits can be considered.

[0141] Step 7: Select the optimal irrigation regime from the Pareto optimal solution set using methods such as the entropy weight method. The optimal method is not limited to the entropy weight method; other methods can be used. In this embodiment, considering the ease of operation of furrow irrigation, the optimal irrigation regime is scheme 3.

[0142] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for determining the optimal irrigation regime on a daily scale based on the coupling of Jensen model, dynamic programming, and crop growth model, characterized in that, include: Step 1: Obtain meteorological, soil, crop parameters, and field trial data for the study area; Step 2: Process the data to obtain the daily input data for the dynamic programming model: daily effective rainfall, daily planned wetting layer depth, daily effective rainfall of the planned wetting layer, daily water sensitivity index of crops, and daily maximum evapotranspiration of crops; Step 3: Develop a daily-scale irrigation regime optimization program based on the Jensen model and dynamic programming algorithm to calculate the optimal daily-scale irrigation scheme based on the Jensen model under water-limited conditions; Step 4: Further adjust the irrigation scheme based on the constraint of minimum irrigation volume per cycle to obtain a better solution set; Step 5: Calibrate and validate the crop growth model based on the experimental data; Step 6: Input the optimal solution set into the calibrated and verified crop growth model to calculate crop yield and water use efficiency, and obtain the Pareto optimal solution set; Step 7: Select and determine the optimal irrigation regime from the Pareto optimal solution set by combining entropy weight method and other methods.

2. The optimization method according to claim 1, characterized in that, Step 1 includes acquiring relevant data such as crop yield, actual water consumption, daily planned wetting layer depth of crops (which can be determined based on local experience), soil bulk density, field water holding capacity, wilting coefficient, and suitable soil moisture content range for crops in a certain research year at the experimental station; and acquiring meteorological data such as daily temperature, rainfall, wind speed, and sunshine duration in the research area in a certain research year.

3. The optimization method according to claim 1, characterized in that, Step 2 processes the data to obtain the daily input data for the dynamic programming model: daily effective rainfall, daily planned wetting layer depth, daily effective rainfall of the planned wetting layer, daily water sensitivity index of the crop, and daily maximum evapotranspiration of the crop.

4. The optimization method according to claim 1, characterized in that, Step 3 involves constructing a daily-scale irrigation regime optimization model based on the Jensen model and dynamic programming algorithm, including: determining state variables; determining decision variables; establishing state transition equations; determining constraints; establishing an objective function; and solving the model using a dimensionality-reduction successive approximation method. First, the soil moisture content is initially estimated, the available water supply is discretized, and the actual evapotranspiration corresponding to different irrigation volumes is calculated in reverse. The initial irrigation volume allocation is determined based on the principle of maximizing cumulative benefits. Then, based on this allocation, the dynamic changes in soil moisture content are calculated in forward direction, and the evapotranspiration is adjusted. The calculated soil moisture content is compared with the initial value, the initial value is modified, and the calculation is repeated until the absolute error between the calculated and initial estimates of the moisture content is less than 0.0001. Finally, a dynamic programming program is developed using Matlab or Visual Studio software to solve the above model.

5. The optimization method according to claim 1, characterized in that, Step 4 includes determining the minimum irrigation volume per application based on different irrigation methods (furrow irrigation, sprinkler irrigation, drip irrigation). Adjust the daily irrigation volume calculation results; if the calculated value for a certain day If the amount of water used is less than the minimum irrigation volume for a certain irrigation method, then the first... i The sum of the irrigation amounts for the day and the next 5 days is obtained. ,like Then the first i The daily irrigation volume is Otherwise the i The daily irrigation amount is set to 0; this yields a better solution set.

6. The optimization method according to claim 1, characterized in that, Step 5: Calibrate and verify the crop growth model based on the experimental data.

7. The optimization method according to claim 1, characterized in that, Step 6: Substitute the better solution set irrigation scheme obtained in step 4 into the AquaCrop model after calibration and verification in step 5, calculate crop yield and water use efficiency, and analyze to obtain the Pareto optimal solution set; the objective function can be single-objective or multi-objective; multi-objectives are not limited to yield and water use efficiency, and other objectives such as economic benefits can be considered.

8. The optimization method according to claim 1, characterized in that, Step 7: Select the optimal irrigation system from the Pareto optimal solution set using methods such as entropy weighting; it is not limited to entropy weighting, other methods can be used for selection.

Images