Intelligent irrigation method, device, equipment, medium and product

Abstract

The present application discloses an intelligent irrigation method, device, equipment, medium and product, which relate to the field of data processing technology. The method comprises obtaining the target yield, soil moisture change and predicted rainfall of a target tea garden in the current cycle; inputting the target yield into a pre-built yield prediction model to obtain predicted evapotranspiration; using the soil moisture change and predicted rainfall to perform calculations to obtain the current irrigation amounts corresponding to various irrigation types; based on the target yield and each predicted evapotranspiration, calculating the fresh leaf level water use efficiency corresponding to each irrigation type; based on the target yield and each current irrigation amount, calculating the economic income data corresponding to each irrigation type; performing normalized dual-objective optimization on the fresh leaf level water use efficiency and the economic income data, respectively, to determine the target irrigation type and the target irrigation amount. The present application can improve the utilization rate of water resources and increase tea yield.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to an intelligent irrigation method, device, equipment, medium and product. Background Art

[0002] As a typical C3 plant, tea trees are extremely sensitive to water conditions. Their photosynthetic rate is significantly limited by stomatal conductance, requiring relatively ample soil moisture to maintain high photosynthetic efficiency and fresh leaf growth rates. Water stress not only reduces the accumulation of photosynthetic products in tea trees but also leads to a decrease in the number of buds and leaves, and a reduction in quality components. Therefore, tea plantations urgently need to dynamically regulate water supply according to the growth period to ensure simultaneous improvement in yield and quality.

[0003] In current tea garden production management, irrigation rates are often set based on farmer experience, timed and quantitative irrigation schedules, or simple meteorological indicators, lacking a precise response to the crop's actual water requirements. In high-latitude tea-growing regions, particularly due to uneven spatial and temporal rainfall distribution, intense evapotranspiration, and insufficient soil water retention, problems often arise, including inefficient irrigation water use, large fluctuations in crop yields, and a disconnect between water management and crop growth. Compared to forest ecosystems, soil moisture dynamics in tea gardens are more chaotic. Influenced by human irrigation intervention, sudden rainfall changes, and evaporation from exposed land surfaces, soil moisture exhibits greater spatial and temporal heterogeneity, with the magnitude and frequency of soil moisture fluctuations significantly greater than in natural vegetation systems. This characteristic exacerbates the complexity of water management in tea gardens, making traditional single-use irrigation systems incapable of responding to rapidly changing water supply and demand balances. Furthermore, the growth characteristics of tea trees differ significantly from those of most grain and vegetable crops. Tea gardens are characterized by multiple harvests, long harvesting periods, and periodic bud bursts, often experiencing spring, summer, and autumn harvests, or even multiple rounds of harvest within a single growing season. Compared to crops that mature once and are harvested over a short period of time (such as rice and wheat), tea trees have a longer water requirement cycle, with water demands fluctuating periodically during the different picking periods. Furthermore, unstable water supply can easily lead to reduced bud and leaf yields and deteriorating quality. Therefore, tea garden irrigation management must not only meet basic physiological water needs but also dynamically adapt to the changing water demands of different growth stages. This places higher demands on the precision and intelligent control of irrigation strategies.

[0004] The current irrigation method only regulates evapotranspiration or soil moisture content, and lacks consideration of the comprehensive optimization of yield, water input and utilization efficiency, resulting in poor water resource utilization and low tea yield. Summary of the Invention

[0005] The purpose of this application is to provide an intelligent irrigation method, device, equipment, medium and product that can improve the utilization rate of water resources and increase tea production.

[0006] To achieve the above objectives, this application provides the following solutions:

[0007] In a first aspect, the present application provides an intelligent irrigation method, comprising:

[0008] Obtain the target yield, soil moisture change, and predicted rainfall of the target tea garden in the current cycle;

[0009] Inputting the target yield into a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to various irrigation types; wherein the irrigation types include drip irrigation, micro-sprinkler irrigation, drip irrigation-mulching, and micro-sprinkler irrigation-mulching;

[0010] Calculate each predicted evapotranspiration amount with the soil moisture change and the predicted rainfall to obtain the current irrigation amount corresponding to each irrigation type;

[0011] Based on the target yield and each predicted evapotranspiration, the fresh leaf level water use efficiency corresponding to each irrigation type is calculated;

[0012] Based on the target yield and each current irrigation amount, calculating economic income data corresponding to each irrigation type;

[0013] Normalized dual-objective optimization is performed on the fresh leaf level water use efficiency and economic income data corresponding to various irrigation types to determine the target irrigation type and the target irrigation amount corresponding to the target irrigation type.

[0014] Optionally, the production forecast model is constructed in the following manner:

[0015] Acquire training data; wherein the training data includes historical irrigation data of the test plots irrigated using various irrigation types, the historical irrigation data including historical rainfall, historical irrigation volume, historical soil moisture content, historical yield, and historical evapotranspiration;

[0016] Based on the historical irrigation data corresponding to various irrigation types, construct the initial prediction models corresponding to various irrigation types;

[0017] Calculate the residuals of each initial prediction model;

[0018] Based on the residuals of each initial prediction model, a residual prediction model of each initial prediction model is established;

[0019] A yield prediction model is constructed based on each initial prediction model and the residual prediction model of each initial prediction model; wherein the yield prediction model includes yield prediction sub-models corresponding to various irrigation types.

[0020] Optionally, the target yield is input into a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to various irrigation types, specifically including:

[0021] Inputting the target yield into a first yield prediction sub-model corresponding to the drip irrigation type in a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to the drip irrigation type;

[0022] Inputting the target yield into a second yield prediction sub-model corresponding to the micro-sprinkler irrigation type in a pre-built yield prediction model to obtain a predicted evapotranspiration corresponding to the micro-sprinkler irrigation type;

[0023] Inputting the target yield into a third yield prediction sub-model corresponding to the drip irrigation-mulching type in a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to the drip irrigation-mulching type;

[0024] The target yield is input into a fourth yield prediction sub-model corresponding to the micro-sprinkler irrigation-mulching type in a pre-constructed yield prediction model to obtain a predicted evapotranspiration corresponding to the micro-sprinkler irrigation-mulching type.

[0025] Optionally, the economic income data corresponding to various irrigation types are calculated based on the target yield and each current irrigation amount, specifically including:

[0026] Obtaining the market price per unit output, irrigation cost per unit area, irrigation water price, and irrigation area of ​​the target tea garden;

[0027] Based on the market price per unit yield, the irrigation cost per unit area, the irrigation water price, the irrigation area, the target yield and each current irrigation amount, economic income data corresponding to each irrigation type is calculated.

[0028] Optionally, the calculation formula for the economic income data is:

[0029] ;

[0030] Among them, EI represents the economic income data corresponding to any irrigation type. represents the target yield, represents the market price of the unit output, represents the irrigation cost per unit area, A represents the irrigation area, represents the irrigation water price, Indicates the current irrigation amount corresponding to any of the irrigation types.

[0031] Optionally, the normalized dual-objective optimization is performed on the fresh leaf level water use efficiency and economic income data corresponding to each irrigation type to determine the target irrigation type and the target irrigation amount corresponding to the target irrigation type, specifically including:

[0032] The water use efficiency at the fresh leaf level and the economic income data corresponding to each irrigation type were normalized to obtain the normalized water use efficiency at the fresh leaf level and the normalized economic income data corresponding to each irrigation type;

[0033] Determining a preset first weight and a second weight, wherein the sum of the first weight and the second weight is 1;

[0034] Based on the first weight and the second weight, normalized fresh leaf level water use efficiency and normalized economic income data corresponding to each irrigation type are calculated to obtain comprehensive benefit values ​​corresponding to each irrigation type;

[0035] The irrigation type corresponding to the maximum comprehensive benefit value is determined as the target irrigation type;

[0036] The current irrigation amount corresponding to the target irrigation type is determined as the target irrigation amount.

[0037] In a second aspect, the present application provides an intelligent irrigation device, comprising:

[0038] An acquisition unit is used to obtain the target yield, soil moisture change and predicted rainfall of the target tea garden in the current cycle;

[0039] An input unit is used to input the target yield into a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to various irrigation types; wherein the irrigation types include drip irrigation, micro-sprinkler irrigation, drip irrigation-mulching, and micro-sprinkler irrigation-mulching;

[0040] a first calculation unit, configured to calculate using each predicted evapotranspiration amount, the soil moisture change amount, and the predicted rainfall amount, respectively, to obtain current irrigation amounts corresponding to various irrigation types;

[0041] A second calculation unit is configured to calculate the fresh leaf level water use efficiency corresponding to each irrigation type based on the target yield and each predicted evapotranspiration;

[0042] A third calculation unit is configured to calculate economic income data corresponding to various irrigation types based on the target yield and each current irrigation amount;

[0043] The optimization unit is used to perform normalized dual-objective optimization on the fresh leaf level water use efficiency and economic income data corresponding to various irrigation types, and determine the target irrigation type and the target irrigation amount corresponding to the target irrigation type.

[0044] In a third aspect, the present application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any one of the above-described intelligent irrigation methods.

[0045] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the above-mentioned intelligent irrigation methods.

[0046] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any one of the above-mentioned intelligent irrigation methods.

[0047] In a sixth aspect, the present application provides a chip, which includes a processor and a communication interface, the communication interface and the processor are coupled, the processor is used to run a program or instruction, and when the processor executes the program or instruction, the steps of any one of the above-mentioned intelligent irrigation methods are implemented.

[0048] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0049] The present application provides an intelligent irrigation method, device, equipment, medium and product. By comprehensively considering multiple factors such as the target yield of the target tea garden, the change in soil moisture and the predicted rainfall, the predicted evaporation corresponding to different irrigation types is accurately obtained with the help of the yield prediction model, and the current irrigation amount is scientifically calculated based on this. At the same time, the water use efficiency and economic income data of the fresh leaf level of different irrigation types are further analyzed in combination with the target yield, and finally the optimal target irrigation type and target irrigation amount are determined through normalized dual-objective optimization. This precise and scientific irrigation decision-making method can reasonably allocate water resources according to the actual needs of the tea garden, effectively avoid the waste of water resources, improve the utilization rate of water resources, and create more suitable moisture conditions for the growth of tea trees through precision irrigation, thereby promoting tea growth and increasing tea yield. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0051] Figure 1 A schematic diagram of a flow chart of an intelligent irrigation method provided in one embodiment of the present application;

[0052] Figure 2 A schematic diagram of the relationship between a yield prediction model and input data provided in one embodiment of the present application;

[0053] Figure 3 A schematic diagram of the fitting results of a yield prediction model provided in one embodiment of the present application;

[0054] Figure 4 A schematic diagram of an irrigation decision result provided in one embodiment of the present application;

[0055] Figure 5 An objective function response spectrum under different irrigation treatment conditions provided in one embodiment of the present application;

[0056] Figure 6 A schematic diagram of the functional modules of an intelligent irrigation device provided in one embodiment of the present application;

[0057] Figure 7 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0059] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0060] In an exemplary embodiment, Figure 1 As shown, a smart irrigation method is provided, which is executed by a computer device, specifically a terminal or a server, etc., and can also be executed by the terminal and the server together. In the embodiment of the present application, the method includes the following steps 101 to 106. Among them:

[0061] Step 101: Obtain the target yield, soil moisture change, and predicted rainfall of the target tea garden in the current cycle.

[0062] Please also refer to Figures 2 to 5 , Figure 2 A schematic diagram of the relationship between a yield prediction model and input data provided in one embodiment of the present application; Figure 3 A schematic diagram of the fitting results of a yield prediction model provided in one embodiment of the present application; Figure 4 A schematic diagram of an irrigation decision result provided in one embodiment of the present application; Figure 5 This is an objective function response graph under different irrigation treatment conditions provided by an embodiment of the present application.

[0063] In the present embodiment, soil moisture variation (ΔW) is a key factor influencing irrigation demand. Soil moisture variation represents the increase or decrease in soil moisture, reflecting the water deficit. During tea plantation growth, soil moisture levels fluctuate due to factors such as evapotranspiration, rainfall, and irrigation. To more accurately predict irrigation demand, the present invention supplements water balance calculations by considering soil moisture variation.

[0064] Step 102: input the target yield into a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to various irrigation types.

[0065] In the embodiment of the present application, the irrigation types include drip irrigation type, micro sprinkler irrigation type, drip irrigation-covering type and micro sprinkler irrigation-covering type.

[0066] For example, experimental area A was selected to carry out different micro-irrigation mode treatment experiments in the three growth stages of spring tea, summer tea and autumn tea from 2021 to 2023. Typical irrigation types such as drip irrigation type, micro-sprinkler irrigation type, drip irrigation-covering type and micro-sprinkler irrigation-covering type were set up respectively. The system collected key indicator data such as irrigation amount, evaporation, yield and soil moisture changes in each stage.

[0067] Among all types of irrigation methods, micro-sprinkler irrigation has good automatic control characteristics, local water supply capabilities and water-saving effects. The system responds quickly and can adjust the water supply frequency and water volume in real time according to the recommended irrigation amount output by the model.

[0068] In addition, the micro-sprinkler irrigation type is easy to integrate with intelligent nodes such as soil moisture sensors, meteorological modules, and control terminals to build a closed-loop control system of "monitoring-decision-making-execution-feedback". It can realize dynamic perception of soil and climate changes and real-time adjustment of irrigation strategies. It is an ideal physical carrier for achieving the long-term optimization and iteration goals of the present invention.

[0069] The soil texture of the experimental plot was brown loam, with an average bulk density of 1.37 g / cm³ at a depth of 0–50 cm and an average field water holding capacity of 24.26% (volume water content). Irrigation methods mainly included drip irrigation, micro-sprinkler irrigation, drip irrigation with mulching, and micro-sprinkler irrigation with mulching. Specific treatment methods are detailed in Table 1:

[0070] Table 1 Experimental treatment table

[0071]

[0072] As an optional implementation, the production forecast model is constructed as follows:

[0073] Acquire training data; wherein the training data includes historical irrigation data of the test plots irrigated using various irrigation types, the historical irrigation data including historical rainfall, historical irrigation volume, historical soil moisture content, historical yield, and historical evapotranspiration;

[0074] Based on the historical irrigation data corresponding to various irrigation types, construct the initial prediction models corresponding to various irrigation types;

[0075] Calculate the residuals of each initial prediction model;

[0076] Based on the residuals of each initial prediction model, a residual prediction model of each initial prediction model is established;

[0077] A yield prediction model is constructed based on each initial prediction model and the residual prediction model of each initial prediction model; wherein the yield prediction model includes yield prediction sub-models corresponding to various irrigation types.

[0078] This implementation method, by constructing initial prediction models for different irrigation types, fully accounts for the unique crop growth environments under each irrigation method, making the model more targeted. Furthermore, the introduction of residual analysis and residual prediction models effectively captures and corrects the prediction bias of the initial model, significantly improving the accuracy and reliability of yield predictions. The resulting yield prediction model, which includes yield prediction sub-models for multiple irrigation types, can provide scientific and accurate yield prediction guidance for agricultural production under different irrigation scenarios, helping agricultural producers rationally plan resource inputs, optimize irrigation strategies, and improve agricultural production efficiency.

[0079] In the embodiment of the present application, historical rainfall can be recorded by a rain gauge;

[0080] The historical irrigation volume can be calculated by multiplying the irrigation intensity of the irrigator by the irrigation time;

[0081] The historical soil moisture content can be measured using Time Domain Reflectometry (TDR), sampling every 10 cm layer, and measuring three random points in each treatment. Due to the shallow root system of tea trees, the measurement depth is 50 cm;

[0082] Tea yield: For each treatment, plants with uniform growth outside the edge of the row were randomly selected for yield measurement. Tea leaves 0.5–1 m apart were picked at each point, and their dry weight was measured (accuracy ≥ 0.01 g) and converted into yield per unit area (kg / hm²).

[0083] Evapotranspiration ET can be calculated based on the soil water balance equation, ignoring the effects of groundwater and surface runoff. The formula is: ,in, is the soil water storage at the beginning of the time period (mm), is the soil water storage at the end of the time period (mm), D is the deep seepage volume, and R is the surface runoff volume.

[0084] Based on the collected historical rainfall P, historical irrigation volume I, historical yield Y, and historical evapotranspiration ET, an initial prediction model was built for the collected evapotranspiration and yield data to capture the major trends in seasonal changes in fresh tea leaf yield.

[0085] Historical irrigation data collection methods and indicators include:

[0086] Rainfall P, recorded daily by setting up an automatic rain gauge or meteorological observation station (unit: mm);

[0087] Irrigation volume I, obtained through irrigation system metering equipment (such as flow meters, irrigation records) irrigation volume data (unit: mm);

[0088] The soil water storage capacity W is calculated from the soil volume water content monitored by sensors at different soil layers. If the soil profile is divided into n layers, the total water storage capacity can be calculated as:

[0089]

[0090] in, is the volumetric water content of the i-th layer of soil, in cm³ / cm³, is the thickness of the soil layer in cm, ρ is the unit conversion factor, which is 10, indicating that 1 cm water layer = 10 mm;

[0091] Evapotranspiration (ET): directly measure ET using a field lysimeter; or indirectly estimate ET based on changes in soil moisture. The calculation formula is:

[0092]

[0093] in, is the soil water storage at the beginning of the time period (mm), is the soil water storage at the end of the period (mm), D is the deep seepage (can be ignored or estimated), and R is the surface runoff (if any, can be deducted).

[0094] Reference crop evapotranspiration and crop coefficient The estimation method can be expressed as follows:

[0095]

[0096] in, It can be calculated according to the Penman-Monteith equation:

[0097]

[0098] in, is the net radiation (MJ / m² / day), G is the soil heat flux (MJ / m² / day), T is the average temperature (℃), is the wind speed at 2 meters height (m / s), 、 is the saturated and actual water vapor pressure (kPa), is the slope of the saturated water vapor pressure curve (kPa / ℃), γ is the dry-wet constant (kPa / ℃), The value is set according to different seasons (spring tea, summer tea, autumn tea), and the typical value range is 0.5–1.2.

[0099] Furthermore, the initial prediction model for D processing is , prediction accuracy =0.784, RMSE (error) =158.14 ;

[0100] The initial prediction model for W processing is , prediction accuracy =0.788, RMSE=139.01 ;

[0101] The initial prediction model for SD treatment is , prediction accuracy =0.731, RMSE=170.93 ;

[0102] The initial prediction model for SW processing is , prediction accuracy =0.877, RMSE=141.66 .

[0103] Furthermore, the initial prediction model part uses random forest regression modeling to compensate for local nonlinear errors and form a cascade prediction model.

[0104] Furthermore, the evapotranspiration (ET) and corresponding yield (Y) data under different treatment conditions were extracted to construct a training set: ;

[0105] Further prediction values ​​are obtained based on the initial prediction model: , where a, b, and c are all preset parameters.

[0106] Furthermore, the residual is calculated: ;in, represents the yield under treatment condition i.

[0107] Furthermore, we use the random forest regressor to For input, It is the target variable, capturing the local nonlinear laws that the initial prediction model fails to fit, such as fluctuations and sudden increases in a small range. Specifically, the residual prediction model is trained: ;

[0108] Furthermore, the two parts of the model are combined in series, that is, the main trend + local compensation, to improve the overall simulation accuracy. The final prediction value is: ;

[0109] From Table 2 (initial prediction model), it can be seen that the prediction accuracy of yield and evapotranspiration under each treatment is relatively low. ) generally ranged from 0.73 to 0.88, with large RMSE prediction errors (up to 170.93). Although the SW treatment achieved the highest ideal yield (2717.22 kg·hm⁻²), its evapotranspiration was also significantly higher (1294.67 mm). This suggests that the traditional method suffers from excessive water use and imprecise regulation in high-yield treatments, making it difficult to achieve the dual goals of yield and water conservation.

[0110] Table 2 Comparison of seasonal optimal production conditions and prediction accuracy

[0111]

[0112] In contrast, Table 3 (yield prediction model of the present invention) shows significantly superior results. All reached above 0.95, RMSE was significantly reduced to 66.99–75.69, and the prediction stability was significantly improved. At the same time, the D, W, and SD treatments reduced the total evapotranspiration (an average decrease of about 50 mm) while increasing the yield (the increase ranged from 90–130 kg·hm⁻²), which fully verified the significant advantages of the method of the present invention in improving prediction accuracy, optimizing water regulation, and improving water use efficiency. In particular, the SW treatment reduced the evapotranspiration from 1294.67 mm to 786.90 mm, while the ideal optimal yield only had a relatively moderate decrease (2432.96 kg·hm⁻²), and the prediction accuracy was 1.3477 kg·hm⁻². It is improved to 0.969, indicating that the algorithm of the present invention can achieve effective water saving under the premise of high yield, has strong adaptability, and has wide promotion value.

[0113] Table 3 Comparison of seasonal optimal production conditions and prediction accuracy of the series model

[0114]

[0115] The present invention demonstrates better model fitting ability and water-saving and yield-increasing effects under different irrigation modes, significantly improving the intelligent decision-making level of the irrigation system.

[0116] As an optional implementation, step 102 inputs the target yield into a pre-built yield prediction model to obtain the predicted evapotranspiration corresponding to various irrigation types, which may include:

[0117] Inputting the target yield into a first yield prediction sub-model corresponding to the drip irrigation type in a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to the drip irrigation type;

[0118] Inputting the target yield into a second yield prediction sub-model corresponding to the micro-sprinkler irrigation type in a pre-built yield prediction model to obtain a predicted evapotranspiration corresponding to the micro-sprinkler irrigation type;

[0119] Inputting the target yield into a third yield prediction sub-model corresponding to the drip irrigation-mulching type in a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to the drip irrigation-mulching type;

[0120] The target yield is input into a fourth yield prediction sub-model corresponding to the micro-sprinkler irrigation-mulching type in a pre-constructed yield prediction model to obtain a predicted evapotranspiration corresponding to the micro-sprinkler irrigation-mulching type.

[0121] This implementation method accurately obtains the predicted evapotranspiration for each irrigation type by inputting the target yield into the yield prediction sub-models pre-built for each irrigation type (drip irrigation, micro-sprinkler irrigation, drip irrigation with mulching, and micro-sprinkler irrigation with mulching). This refined processing fully considers the unique impact of different irrigation methods and their combinations on crop evapotranspiration, making the prediction results more targeted and accurate. It provides a strong basis for agricultural producers to rationally plan water resources and develop scientific irrigation plans under different irrigation scenarios, helping to improve water resource utilization efficiency and promote the sustainable development of agricultural production.

[0122] Step 103 : Calculate each predicted evapotranspiration amount with the soil moisture change and the predicted rainfall to obtain the current irrigation amount corresponding to each irrigation type.

[0123] In the embodiment of the present application, when considering the calculation of irrigation amount, the water balance formula is added:

[0124]

[0125] Where I is the current irrigation amount, ET is the predicted evapotranspiration, P is the predicted rainfall, and ΔW is the soil moisture change, which is obtained by the soil moisture sensor.

[0126] In this implementation, we assume a ΔW of -10 mm, indicating a moderate water deficit for the crop at this stage. Under a fixed rainfall of P = 135.6 mm and a predicted yield of ≥ 1500 kg / hm², the yield prediction model is used to reversely calculate the required evapotranspiration and irrigation volume to determine the optimal water management strategy. In this case, we assume a ΔW of -10 mm, indicating that the tea trees are experiencing a water deficit.

[0127] The required evapotranspiration (ET) for different treatments can be calculated based on the target yield range and rainfall, and the appropriate irrigation amount (I) can be determined accordingly.

[0128] Figure 4 The coupling relationship between yield and evapotranspiration under four irrigation types is shown. Each data point in the figure represents a corresponding combination of yield and evapotranspiration, and the size of different points indicates the recommended irrigation amount output by the model, that is, the irrigation demand obtained by reverse deduction based on the yield prediction model. The figure visually expresses the logical path and decision-making relationship between yield targets, evapotranspiration process and irrigation water volume, providing an intuitive basis for formulating water-saving and efficient precision irrigation plans. According to the existing water source conditions and the yield-evapotranspiration-irrigation volume relationship presented in the figure, users can choose the irrigation plan that best meets actual needs from a variety of possible combinations to achieve scientific allocation of water resources and yield optimization. In addition, Figure 4It also has strong practical guiding significance and can be used as an auxiliary decision-making tool for field irrigation management, helping users to formulate flexible and reasonable dynamic irrigation plans under weather changes and resource constraints, and improve the intelligence level of agricultural water resources management.

[0129] Step 104 : Based on the target yield and each predicted evapotranspiration, calculate the fresh leaf level water use efficiency corresponding to each irrigation type.

[0130] In the embodiment of the present application, the calculation formula for calculating the water use efficiency (WUE) of fresh leaves is:

[0131]

[0132] Where Y represents the target yield and ET represents the predicted evapotranspiration.

[0133] Step 105: Based on the target yield and each current irrigation amount, calculate and obtain economic income data corresponding to each irrigation type.

[0134] As an optional implementation, step 105 may include calculating the economic income data corresponding to various irrigation types based on the target yield and the current irrigation amounts:

[0135] Obtaining the market price per unit output, irrigation cost per unit area, irrigation water price, and irrigation area of ​​the target tea garden;

[0136] Based on the market price per unit yield, the irrigation cost per unit area, the irrigation water price, the irrigation area, the target yield and each current irrigation amount, economic income data corresponding to each irrigation type is calculated.

[0137] This method, which comprehensively obtains key information on the target tea garden's unit yield market price, unit irrigation cost, irrigation water price, and irrigated area, and then calculates economic income data corresponding to different irrigation types based on this information, combined with the target yield and current irrigation volume, can comprehensively and accurately consider the cost input and benefit output of different irrigation methods. This calculation method fully considers various economic factors in actual production, providing tea garden managers with a clear and accurate basis for economic benefit evaluation, helping them make scientific and reasonable choices between different irrigation types, optimize resource allocation, effectively control costs while ensuring tea production, maximize economic benefits, and promote the sustainable development of tea gardens.

[0138] The calculation formula for economic income data is:

[0139] ;

[0140] Among them, EI represents the economic income data corresponding to any irrigation type. represents the target yield, represents the market price of the unit output, represents the irrigation cost per unit area, A represents the irrigation area, represents the irrigation water price, Indicates the current irrigation amount corresponding to any of the irrigation types.

[0141] Step 106 , performing normalized dual-objective optimization on the fresh leaf level water use efficiency and economic income data corresponding to various irrigation types, and determining a target irrigation type and a target irrigation amount corresponding to the target irrigation type.

[0142] As an optional implementation, step 106 performs normalized dual-objective optimization on the fresh leaf water use efficiency and economic income data corresponding to each irrigation type. The method for determining the target irrigation type and the target irrigation amount corresponding to the target irrigation type may include:

[0143] The water use efficiency at the fresh leaf level and the economic income data corresponding to each irrigation type were normalized to obtain the normalized water use efficiency at the fresh leaf level and the normalized economic income data corresponding to each irrigation type;

[0144] Determining a preset first weight and a second weight, wherein the sum of the first weight and the second weight is 1;

[0145] Based on the first weight and the second weight, normalized fresh leaf level water use efficiency and normalized economic income data corresponding to each irrigation type are calculated to obtain comprehensive benefit values ​​corresponding to each irrigation type;

[0146] The irrigation type corresponding to the maximum comprehensive benefit value is determined as the target irrigation type;

[0147] The current irrigation amount corresponding to the target irrigation type is determined as the target irrigation amount.

[0148] Among them, the implementation of this implementation method, by normalizing the water use efficiency and economic income data at the fresh leaf level corresponding to different irrigation types, eliminates the influence of data dimension differences and ensures the comparability of the two in the optimization process. Combined with the first and second weights that are pre-set and sum to 1, the two key indicators of water use efficiency and economic income are organically integrated to calculate the comprehensive benefit value, which comprehensively considers the water-saving benefits and economic benefits in irrigation decisions. Finally, the target irrigation type and the corresponding target irrigation amount are determined based on the maximum comprehensive benefit value. This dual-objective optimization method can provide a scientific, reasonable and practical basis for tea garden irrigation decisions, helping tea gardens to effectively improve water resource utilization efficiency while ensuring economic benefits, achieve the optimal balance between water resources and economic benefits, and promote efficient and sustainable development of tea garden production.

[0149] In the embodiment of the present application, the calculation formula for normalizing the fresh leaf level water use efficiency (WUE) and economic income data (EI) of multiple micro-irrigation treatments is:

[0150]

[0151]

[0152] in, To normalize economic income data, is the normalized water use efficiency at the fresh leaf level, For economic income data under different treatments, For each processing maximum economic income data, Each process has the minimum economic income data, is the maximum water use efficiency of each treatment, is the minimum water use efficiency of each treatment.

[0153] Furthermore, the calculation formula for the comprehensive benefit value F corresponding to any irrigation type can be:

[0154]

[0155] Among them, α is the first weight, β is the second weight, and α+β=1. The values ​​of α and β can be flexibly set according to user control preferences.

[0156] Assuming that the weights of water use efficiency and economic income are both 0.5, the calculation formula for the comprehensive benefit value F can be: .

[0157] Furthermore, the irrigation output is optimized to maximize the comprehensive objective function F, and the recommended evapotranspiration demand and corresponding irrigation strategy are reversed to ultimately achieve intelligent irrigation regulation.

[0158] Further, Figure 5 The objective function response spectrum under different treatment conditions calculated according to the present invention shows that the overall comprehensive benefit trends of D (drip irrigation type), W (micro-sprinkler irrigation type) and SD (drip irrigation-mulching type) treatments are similar, and there are certain differences in SW (micro-sprinkler irrigation-mulching type) treatment. However, all treatments have optimal control strategies at different levels, and all treatments have clear optimal irrigation volume ranges, providing managers with a variety of control path options and greater flexibility and adaptability.

[0159] During the process of executing irrigation regulation, the present invention sets up a long-term data monitoring and iterative optimization mechanism to improve the intelligence and precision level of tea garden irrigation.

[0160] First, data recording and error analysis, in each irrigation cycle, record the actual evapotranspiration , actual output , actual irrigation volume , actual rainfall , and soil moisture changes At the same time, the forecast value is calculated through the existing yield forecast model, and the forecast error is evaluated.

[0161] Furthermore, the recorded data are used to update the regression model and random forest model parameters to adapt them to the current climate, moisture or management change trends and improve the model generalization ability.

[0162] Furthermore, using the updated model, the latest meteorological forecast data (such as rainfall P) and predicted evapotranspiration are input. , planned target output , and soil moisture estimation , calculate the optimal irrigation amount: .

[0163] Furthermore, steps 101-103 form a closed-loop feedback system, which is continuously performed in each cycle, gradually converging to the optimal irrigation strategy. As data accumulates, the model's prediction accuracy continues to improve, supporting more efficient irrigation regulation and yield prediction.

[0164] In the implementation of this invention, the core of step 1034 is long-term data monitoring of irrigation volume and continuous optimization of irrigation plans through an iterative process. During each irrigation cycle, the actual irrigation volume and evapotranspiration (ET) are recorded for each period based on the inferred irrigation volume (I). This data is then combined with soil moisture changes (ΔW) and rainfall (P) in the tea garden for data monitoring. The data feedback after each irrigation cycle serves as input for subsequent iterations, adjusting and optimizing irrigation decisions.

[0165] Furthermore, after each irrigation cycle, the model is revised based on actual monitoring data (such as soil moisture, yield, and evapotranspiration). The adjusted model is then used for prediction and irrigation adjustment in the next cycle. By repeating this process, the system gradually converges, ultimately forming an optimal irrigation prediction model to support long-term water regulation and yield optimization.

[0166] Furthermore, during each iteration, the actual yield for each period is compared with the target yield range to evaluate the effectiveness of the irrigation plan. By comparing simulated and measured data, the accuracy of the system's predictions is evaluated, and appropriate adjustments are made to the model to improve the accuracy of future predictions. This process helps improve the accuracy of irrigation management and avoid over-irrigation or water deficits.

[0167] The above analysis demonstrates that the proposed intelligent tea garden irrigation method, based on yield prediction and dual-objective optimization and control, exhibits strong adaptability and accuracy. By constructing a tandem quadratic regression model and a random forest residual compensation model, it achieves high-precision simulation and prediction of tea garden yield under different irrigation treatments. Combined with target yield constraints, the system can infer the required evapotranspiration and corresponding irrigation volume to meet actual production needs.

[0168] In the patent implementation case, the present invention takes the drip irrigation type, micro-sprinkler irrigation type, drip irrigation-covering type and micro-sprinkler irrigation-covering type in the spring tea stage as examples to construct a yield prediction model between irrigation amount and yield, and based on the set target yield range, the optimal evaporation and transpiration combination is deduced, and then the water deficit ΔW is introduced through the soil water balance principle to accurately determine the optimal irrigation demand under the target conditions, which has strong practicality and scalability.

[0169] Furthermore, this invention incorporates two optimization objectives, water use efficiency (WUE) and economic income (EI), into irrigation control strategies, constructing a unified, comprehensive objective function that addresses the dual needs of water conservation and yield increase. While delivering irrigation decision output, the objective function response graph presents unimodal or bimodal benefit distributions, enabling users to flexibly select optimal irrigation ranges based on actual resource conditions under various irrigation modes, achieving differentiated management.

[0170] As a high-frequency, low-volume, localized water supply irrigation method, the micro-irrigation system has strong control flexibility and real-time response, and is particularly suitable for the intelligent irrigation strategy based on yield prediction and refined water control proposed in the present invention. Compared with traditional flooding or sprinkler irrigation systems, the micro-irrigation system can achieve more precise water supply control through deep coupling with sensor networks and intelligent control units. During the implementation of the present invention, the micro-irrigation system, as the execution layer, can accurately control the irrigation water volume of each tea tree root zone based on the irrigation volume recommendations output in real time by the algorithm model, and dynamically update the control strategy based on feedback data, thereby effectively improving water use efficiency, reducing deep leakage and water waste, and ensuring the target yield is achieved.

[0171] In addition, the micro-irrigation system has strong automation interface capabilities, which facilitates the integration of intelligent hardware such as soil moisture sensors, meteorological modules, and communication modules to form a data closed loop, providing a hardware foundation for the long-term monitoring, dynamic feedback, and model iteration required by the present invention, reflecting the application advantages and adaptability of the system in smart agriculture scenarios. It is worth noting that the above embodiments are only used to illustrate the technical solution of the present invention and are not limiting. Those skilled in the art should understand that the present invention may be modified or equivalently replaced without departing from the purpose and scope of the technical solution of the present invention, and these modifications or replacements should also be included in the scope of protection of the present invention.

[0172] By implementing the above steps 101 to 106, water resources can be reasonably allocated according to the actual needs of the tea garden, effectively avoiding the waste of water resources, improving the utilization rate of water resources, and creating more suitable moisture conditions for the growth of tea trees through precision irrigation, thereby promoting the growth of tea leaves and increasing tea yields. In addition, the present application can also provide scientific and accurate yield prediction guidance for agricultural production under different irrigation scenarios, helping agricultural producers to rationally plan resource inputs, optimize irrigation strategies, and improve agricultural production efficiency. In addition, the present application can also fully consider the unique impact of different irrigation methods and their combinations on the crop evaporation process, making the prediction results more targeted and accurate, providing a strong basis for agricultural producers to rationally plan water resources and formulate scientific irrigation plans under different irrigation scenarios, helping to improve water resource utilization efficiency and promote the sustainable development of agricultural production. In addition, the present application can also make scientific and reasonable choices between different irrigation types, optimize resource allocation, effectively control costs while ensuring tea yield, maximize economic benefits, and promote the sustainable development of tea gardens. In addition, this application can also provide a scientific, reasonable and practical basis for tea garden irrigation decision-making, helping tea gardens to effectively improve water resource utilization efficiency while ensuring economic benefits, achieve the optimal balance between water resources and economic benefits, and promote efficient and sustainable development of tea garden production.

[0173] Based on the same inventive concept, embodiments of the present application also provide an intelligent irrigation device for implementing the aforementioned intelligent irrigation method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more intelligent irrigation device embodiments provided below can be found in the aforementioned limitations of the intelligent irrigation method and will not be further elaborated here.

[0174] In an exemplary embodiment, Figure 6 As shown, an intelligent irrigation device is provided, comprising:

[0175] An acquisition unit 601 is used to acquire the target yield, soil moisture change, and predicted rainfall of the target tea garden in the current cycle;

[0176] Input unit 602 is used to input the target yield into a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to various irrigation types; wherein the irrigation types include drip irrigation, micro-sprinkler irrigation, drip irrigation-mulching, and micro-sprinkler irrigation-mulching;

[0177] A first calculation unit 603 is configured to calculate each predicted evapotranspiration amount with the soil moisture change and the predicted rainfall to obtain a current irrigation amount corresponding to each irrigation type;

[0178] The second calculation unit 604 is configured to calculate the fresh leaf level water use efficiency corresponding to each irrigation type based on the target yield and each predicted evapotranspiration;

[0179] The third calculation unit 605 is configured to calculate economic income data corresponding to various irrigation types based on the target yield and each current irrigation amount;

[0180] The optimization unit 606 is used to perform normalized dual-objective optimization on the fresh leaf level water use efficiency and economic income data corresponding to various irrigation types, and determine the target irrigation type and the target irrigation amount corresponding to the target irrigation type.

[0181] The implementation of the above-mentioned embodiment can reasonably allocate water resources according to the actual needs of the tea garden, effectively avoid the waste of water resources, improve the utilization rate of water resources, and create more suitable moisture conditions for the growth of tea trees through precise irrigation, thereby promoting tea growth and increasing tea production.

[0182] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 7As shown. The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. The processor, memory and input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store intelligent irrigation data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, an intelligent irrigation method is implemented.

[0183] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0184] In an exemplary embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0185] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0186] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0187] In an exemplary embodiment, a chip is provided, which includes a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the steps in the above-mentioned method embodiments and achieve the same technical effects. To avoid repetition, they are not described here.

[0188] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0189] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0190] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0191] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0192] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0193] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. An intelligent irrigation method, characterized in that: The intelligent irrigation method comprises: Obtain the target yield, soil moisture change, and predicted rainfall of the target tea garden in the current cycle; Inputting the target yield into a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to various irrigation types; wherein the irrigation types include drip irrigation, micro-sprinkler irrigation, drip irrigation-mulching, and micro-sprinkler irrigation-mulching; Calculate each predicted evapotranspiration amount with the soil moisture change and the predicted rainfall to obtain the current irrigation amount corresponding to each irrigation type; Based on the target yield and each predicted evapotranspiration, the fresh leaf level water use efficiency corresponding to each irrigation type is calculated; Based on the target yield and each current irrigation amount, calculating economic income data corresponding to each irrigation type; Normalized dual-objective optimization is performed on the fresh leaf level water use efficiency and economic income data corresponding to various irrigation types to determine the target irrigation type and the target irrigation amount corresponding to the target irrigation type; The production forecast model is constructed as follows: Acquire training data; wherein the training data includes historical irrigation data of the test plots irrigated using various irrigation types, the historical irrigation data including historical rainfall, historical irrigation volume, historical soil moisture content, historical yield, and historical evapotranspiration; Based on the historical irrigation data corresponding to various irrigation types, construct the initial prediction models corresponding to various irrigation types; Calculate the residuals of each initial prediction model; Based on the residuals of each initial prediction model, a residual prediction model of each initial prediction model is established; Constructing a yield prediction model based on each initial prediction model and the residual prediction model of each initial prediction model; wherein the yield prediction model includes yield prediction sub-models corresponding to various irrigation types; Furthermore, the target yield is input into a pre-built yield prediction model to obtain the predicted evapotranspiration corresponding to various irrigation types, specifically including: Inputting the target yield into a first yield prediction sub-model corresponding to the drip irrigation type in a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to the drip irrigation type; Inputting the target yield into a second yield prediction sub-model corresponding to the micro-sprinkler irrigation type in a pre-built yield prediction model to obtain a predicted evapotranspiration corresponding to the micro-sprinkler irrigation type; Inputting the target yield into a third yield prediction sub-model corresponding to the drip irrigation-mulching type in a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to the drip irrigation-mulching type; The target yield is input into a fourth yield prediction sub-model corresponding to the micro-sprinkler irrigation-mulching type in a pre-constructed yield prediction model to obtain a predicted evapotranspiration corresponding to the micro-sprinkler irrigation-mulching type.

2. The intelligent irrigation method according to claim 1, characterized in that: The economic income data corresponding to each irrigation type is calculated based on the target yield and each current irrigation amount, specifically including: Obtaining the market price per unit output, irrigation cost per unit area, irrigation water price, and irrigation area of ​​the target tea garden; Based on the market price per unit yield, the irrigation cost per unit area, the irrigation water price, the irrigation area, the target yield and each current irrigation amount, economic income data corresponding to each irrigation type is calculated.

3. The intelligent irrigation method according to claim 2, characterized in that: The calculation formula for the economic income data is: ； Among them, EI represents the economic income data corresponding to any irrigation type. represents the target yield, represents the market price of the unit output, represents the irrigation cost per unit area, A represents the irrigation area, represents the irrigation water price, Indicates the current irrigation amount corresponding to any of the irrigation types.

4. The intelligent irrigation method according to any one of claims 1 to 3, characterized in that: The normalized dual-objective optimization is performed on the fresh leaf level water use efficiency and economic income data corresponding to each irrigation type to determine the target irrigation type and the target irrigation amount corresponding to the target irrigation type, specifically including: The water use efficiency at the fresh leaf level and the economic income data corresponding to each irrigation type were normalized to obtain the normalized water use efficiency at the fresh leaf level and the normalized economic income data corresponding to each irrigation type; Determining a preset first weight and a second weight, wherein the sum of the first weight and the second weight is 1; Based on the first weight and the second weight, normalized fresh leaf level water use efficiency and normalized economic income data corresponding to each irrigation type are calculated to obtain comprehensive benefit values ​​corresponding to each irrigation type; The irrigation type corresponding to the maximum comprehensive benefit value is determined as the target irrigation type; The current irrigation amount corresponding to the target irrigation type is determined as the target irrigation amount.

5. An intelligent irrigation device, characterized in that: The intelligent irrigation device comprises: An acquisition unit is used to obtain the target yield, soil moisture change and predicted rainfall of the target tea garden in the current cycle; An input unit is used to input the target yield into a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to various irrigation types; wherein the irrigation types include drip irrigation, micro-sprinkler irrigation, drip irrigation-mulching, and micro-sprinkler irrigation-mulching; a first calculation unit, configured to calculate using each predicted evapotranspiration amount, the soil moisture change amount, and the predicted rainfall amount, respectively, to obtain current irrigation amounts corresponding to various irrigation types; A second calculation unit is configured to calculate the fresh leaf level water use efficiency corresponding to each irrigation type based on the target yield and each predicted evapotranspiration; A third calculation unit is configured to calculate economic income data corresponding to various irrigation types based on the target yield and each current irrigation amount; an optimization unit for performing normalized dual-objective optimization on fresh leaf level water use efficiency and economic income data corresponding to various irrigation types, and determining a target irrigation type and a target irrigation amount corresponding to the target irrigation type; The production forecast model is constructed as follows: Acquire training data; wherein the training data includes historical irrigation data of the test plots irrigated using various irrigation types, the historical irrigation data including historical rainfall, historical irrigation volume, historical soil moisture content, historical yield, and historical evapotranspiration; Based on the historical irrigation data corresponding to various irrigation types, construct the initial prediction models corresponding to various irrigation types; Calculate the residuals of each initial prediction model; Based on the residuals of each initial prediction model, a residual prediction model of each initial prediction model is established; Constructing a yield prediction model based on each initial prediction model and the residual prediction model of each initial prediction model; wherein the yield prediction model includes yield prediction sub-models corresponding to various irrigation types; Furthermore, the input unit inputs the target yield into a pre-built yield prediction model to obtain the predicted evapotranspiration corresponding to various irrigation types in the following manner: Inputting the target yield into a first yield prediction sub-model corresponding to the drip irrigation type in a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to the drip irrigation type; Inputting the target yield into a second yield prediction sub-model corresponding to the micro-sprinkler irrigation type in a pre-built yield prediction model to obtain a predicted evapotranspiration corresponding to the micro-sprinkler irrigation type; Inputting the target yield into a third yield prediction sub-model corresponding to the drip irrigation-mulching type in a pre-built yield prediction model to obtain predicted evapotranspiration corresponding to the drip irrigation-mulching type; The target yield is input into a fourth yield prediction sub-model corresponding to the micro-sprinkler irrigation-mulching type in a pre-constructed yield prediction model to obtain a predicted evapotranspiration corresponding to the micro-sprinkler irrigation-mulching type.

6. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the intelligent irrigation method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent irrigation method according to any one of claims 1 to 4 are implemented.

8. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the intelligent irrigation method according to any one of claims 1 to 4 are implemented.

Images