Crop irrigation method, system, electronic equipment and medium

Abstract

The present invention discloses a crop irrigation method, system, electronic device, and medium, relating to the field of agricultural technology. The method comprises: training an Aquacrop model using a sample set and a preset irrigation amount; obtaining a first irrigation water prediction amount based on historical meteorological data and crop coefficients; training a dynamic irrigation water prediction model using future meteorological data and historical meteorological data as input and the first irrigation water prediction amount as output; inputting the future meteorological data into the trained dynamic irrigation water prediction model to obtain a second irrigation water prediction amount; optimizing the trained Aquacrop model with the goal of achieving a difference between the preset irrigation amount and the second irrigation water prediction amount of zero; and irrigating using the optimized Aquacrop model with the goal of maximizing crop yield and minimizing the total amount of irrigation water. The present invention improves the accuracy of the Aquacrop model, thereby enabling water planning to meet actual irrigation needs.

Description

Technical Field

[0001] The present invention relates to the field of agricultural technology, and in particular to a crop irrigation method, system, electronic equipment and medium. Background Art

[0002] The core of irrigation water management is planned water use, which is currently divided into static and dynamic water use. Static water use involves pre-compiling a static water use plan based on historical data, and then temporarily revising the plan based on actual conditions. However, in actual production, meteorological factors, crop factors, soil factors, water resource conditions, and canal system conditions within the irrigation area do not fully match long-term monitoring, and water use plans often fail to meet actual irrigation requirements. Dynamic water use involves irrigation using a crop dynamic irrigation water model based on a machine learning algorithm. However, this model is not accurate enough, resulting in water use plans that fail to meet actual irrigation requirements. Summary of the Invention

[0003] The purpose of the present invention is to provide a crop irrigation method, system, electronic device and medium that can improve the accuracy of the Aquacrop model and thus use water planning to meet actual irrigation needs.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] A crop irrigation method comprising:

[0006] Obtaining a sample set and future meteorological data for the site where the crop is located; the sample set includes historical growth data, historical field management data, historical soil parameter data, historical yield data, and historical meteorological data for the site where the crop is located;

[0007] Using the sample set and the preset irrigation amount to train the Aquacrop model to obtain a trained Aquacrop model;

[0008] Obtaining a first predicted amount of irrigation water for the crop based on historical meteorological data of the site where the crop is located and a crop coefficient of the crop;

[0009] Using future meteorological data of the site where the crop is located and historical meteorological data of the site where the crop is located as input and the first predicted amount of irrigation water as output, a deep learning algorithm is used to train a dynamic irrigation water prediction model to obtain a trained dynamic irrigation water prediction model;

[0010] Inputting future meteorological data of the site where the crop is located into the trained dynamic irrigation water prediction model to obtain a second irrigation water prediction amount;

[0011] Taking the difference between the preset irrigation amount of the crop and the second predicted irrigation water amount of the crop as 0 as a target, optimizing the trained Aquacrop model parameters to obtain an optimized Aquacrop model;

[0012] With the goal of maximizing crop yield and minimizing the total amount of irrigation water, the optimized Aquacrop model is used for irrigation.

[0013] Optionally, the sample set and the preset irrigation amount are used to train the Aquacrop model to obtain a trained Aquacrop model, specifically including:

[0014] The Aquacrop model is trained using the preset irrigation amount and historical growth data, historical field management data, historical soil parameter data, and historical meteorological data of the site of the crops in the sample set as input, and using the historical yield of the crops in the sample set as output to obtain a trained Aquacrop model.

[0015] Optionally, obtaining a first predicted amount of irrigation water for the crop based on historical meteorological data of the site where the crop is located and a crop coefficient of the crop specifically includes:

[0016] The evapotranspiration of the reference crop is obtained based on the historical meteorological data of the site where the crop is located and the Penman formula;

[0017] According to the formula W1=k c ET0-P calculates the first predicted amount of irrigation water for crops, where W1 represents the first predicted amount of irrigation water for crops, k c represents the crop coefficient of the crop, ET0 represents the evapotranspiration of the reference crop, and P represents the daily effective rainfall.

[0018] Optionally, the calculation formula for the daily effective rainfall is:

[0019] Among them, P m It represents daily rainfall, and mm represents millimeters.

[0020] A crop irrigation system comprising:

[0021] An acquisition module is used to acquire a sample set and future meteorological data of the site where the crop is located; the sample set includes historical growth data, historical field management data, historical soil parameter data, historical yield data, and historical meteorological data of the site where the crop is located;

[0022] An Aquacrop model training module is used to train the Aquacrop model using the sample set and the preset irrigation amount to obtain a trained Aquacrop model;

[0023] A first irrigation water forecast amount determination module, configured to obtain a first irrigation water forecast amount for the crop based on historical meteorological data of the site where the crop is located and a crop coefficient of the crop;

[0024] a dynamic irrigation water prediction model training module, configured to use future meteorological data of the site where the crop is located and historical meteorological data of the site where the crop is located as input, and the first predicted amount of irrigation water as output, and to train the dynamic irrigation water prediction model using a deep learning algorithm to obtain a trained dynamic irrigation water prediction model;

[0025] A second irrigation water predicted amount determination module is configured to input future meteorological data of the site where the crop is located into the trained dynamic irrigation water prediction model to obtain a second irrigation water predicted amount;

[0026] An Aquacrop model optimization module is used to optimize the trained Aquacrop model parameters to obtain an optimized Aquacrop model with the difference between the preset irrigation amount of the crop and the second predicted irrigation water amount of the crop being 0 as a target;

[0027] The irrigation module is used to irrigate with the optimized Aquacrop model with the goal of maximizing crop yield and minimizing the total amount of irrigation water.

[0028] Optionally, the Aquacrop model training module specifically includes:

[0029] The Aquacrop model training unit is used to train the Aquacrop model using the preset irrigation amount and historical growth data, historical field management data, historical soil parameter data, and historical meteorological data of the site of the crops in the sample set as input, and using the historical yield of the crops in the sample set as output to obtain a trained Aquacrop model.

[0030] Optionally, the first irrigation water pre-measurement determination module specifically includes:

[0031] A reference crop evapotranspiration calculation unit, configured to obtain the reference crop evapotranspiration based on historical meteorological data of the site where the crop is located and the Penman formula;

[0032] The first irrigation water prediction calculation unit is used to calculate the amount of irrigation water according to the formula W1=k c ET0-P calculates the first predicted amount of irrigation water for crops, where W1 represents the first predicted amount of irrigation water for crops, k c represents the crop coefficient of the crop, ET0 represents the evapotranspiration of the reference crop, and P represents the daily effective rainfall.

[0033] Optionally, the calculation formula for the daily effective rainfall is:

[0034] Among them, P m It represents daily rainfall, and mm represents millimeters.

[0035] An electronic device, comprising:

[0036] A memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the crop irrigation method according to the above description.

[0037] A computer-readable storage medium stores a computer program, wherein the computer program implements the above-mentioned crop irrigation method when executed by a processor.

[0038] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0039] The present invention couples the Aquacrop crop model with the crop dynamic irrigation water model, iterates repeatedly until the difference between the irrigation water predictions of the two models is 0, improves the accuracy of the Aquacrop model, obtains the optimal Aquacrop model, and then uses water planning to meet actual irrigation needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention 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 invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0041] Figure 1 A general flow chart of the crop irrigation method provided by an embodiment of the present invention;

[0042] Figure 2 A flowchart of the Aquacrop model optimization provided by an embodiment of the present invention;

[0043] Figure 3 A flow chart of a crop irrigation method provided by an embodiment of the present invention;

[0044] Figure 4 This is a block diagram of the smart irrigation control system based on cloud-edge-end collaboration provided by an embodiment of the present invention;

[0045] Figure 5 A diagram of the microservice architecture provided for an embodiment of the present invention;

[0046] Figure 6A task collaboration relationship diagram provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0048] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0049] An embodiment of the present invention provides a crop irrigation method, comprising:

[0050] Obtain a sample set and future meteorological data of the site where the crop is located; the sample set includes historical growth data, historical field management data, historical soil parameter data, historical yield and historical meteorological data of the site where the crop is located.

[0051] The sample set and the preset irrigation amount are used to train the Aquacrop model to obtain a trained Aquacrop model.

[0052] A first predicted amount of irrigation water for the crop is obtained based on historical meteorological data of the site where the crop is located and a crop coefficient of the crop.

[0053] Taking the future meteorological data of the site where the crop is located and the historical meteorological data of the site where the crop is located as input and the first predicted amount of irrigation water as output, a deep learning algorithm is used to train the dynamic irrigation water prediction model to obtain a trained dynamic irrigation water prediction model.

[0054] The future meteorological data of the site where the crop is located is input into the trained dynamic irrigation water prediction model to obtain a second irrigation water prediction amount.

[0055] Taking the difference between the preset irrigation amount of the crop and the second predicted irrigation water amount of the crop as 0 as a goal, the trained Aquacrop model parameters are optimized to obtain an optimized Aquacrop model.

[0056] With the goal of maximizing crop yield and minimizing the total amount of irrigation water, the optimized Aquacrop model is used for irrigation.

[0057] In practical applications, the sample set and the preset irrigation amount are used to train the Aquacrop model to obtain a trained Aquacrop model, specifically including:

[0058] The Aquacrop model is trained using the preset irrigation amount and historical growth data, historical field management data, historical soil parameter data, and historical meteorological data of the site of the crops in the sample set as input, and using the historical yield of the crops in the sample set as output to obtain a trained Aquacrop model.

[0059] In practical applications, obtaining the first predicted amount of irrigation water for the crop based on historical meteorological data of the site where the crop is located and the crop coefficient of the crop specifically includes:

[0060] The reference crop evapotranspiration is obtained based on the historical meteorological data of the site where the crop is located and the Penman formula.

[0061] According to the formula W1=k c ET0-P calculates the first predicted amount of irrigation water for crops, where W1 represents the first predicted amount of irrigation water for crops, k c represents the crop coefficient of the crop, ET0 represents the evapotranspiration of the reference crop, and P represents the daily effective rainfall.

[0062] In practical applications, the calculation formula for the daily effective rainfall is:

[0063] Among them, P m It represents daily rainfall, and mm represents millimeters.

[0064] The present invention provides a more specific embodiment to introduce the above method in detail, and the specific steps are as follows: Figure 1 and Figure 3 As shown:

[0065] Step 1: Obtain site meteorological data, crop growth data, field management data, soil parameter data, and crop yield data. Collect historical and real-time data on the site's meteorological data, crop growth data (crop variety, growth cycle, planting density, maximum canopy cover, canopy attenuation coefficient, harvest index lifetime, and crop coefficient), field management data (tillage method, film cover, weed removal level, and irrigation schedule (including irrigation volume and timing), soil parameter data (number of soil layers, soil moisture, permanent wilting point, field water holding capacity, saturated moisture content), and crop yield.

[0066] Step 2: Use Aquacrop to simulate total crop irrigation water and calibrate parameters. The Aquacrop model uses the preset irrigation volume (W), i.e., total irrigation water, historical meteorological data, historical crop growth data, historical soil parameter data, and historical field management data, as inputs. The model uses historical crop yields as outputs to calibrate parameters.

[0067] Step 3: Calculate the crop water requirement using the FAO-56 formula combined with the crop coefficient method. Based on the historical meteorological data of the crop site, the reference evapotranspiration ET0 is calculated using the Penman formula proposed by FAO, and the crop coefficient k of the corresponding crop in the corresponding area is used. c (k of different crops at different growth stages c The crop water requirement is calculated (the value is different) and then the crop irrigation water volume W1, i.e. the first irrigation water forecast, is calculated using the formula ET c =k c ET0, and W1=ET c -P, where ET0 represents the reference crop evapotranspiration, in mm / d; △ represents the slope of the water vapor pressure curve, in kPa / ℃; A represents the difference between the net radiation of the crop surface and the soil heat flux density, in MJ / (m 2 d); γ is the hygrometer constant, unit is kPa / ℃; w is the wind speed 2m above the ground, unit is m / s; v is the saturated pressure deficit, unit is kPa; T a Indicates the average temperature 2m above the ground, in °C, ET c Indicates crop water requirement, unit: mm; P m represents daily rainfall in mm, P represents daily effective rainfall in mm, k c Represents the crop coefficient.

[0068] Step 4: Use deep learning algorithms combined with soil moisture and data from the National Meteorological Center to build a dynamic irrigation water prediction model. Based on the historical meteorological data, soil moisture, and future meteorological data of the crop site, use deep learning algorithms to train the dynamic irrigation water prediction model. Input the future meteorological data into the trained dynamic irrigation water prediction model to obtain the second irrigation water prediction amount W2. Figure 2 As shown in the figure, the crop irrigation water volume W simulated by the Aquacrop crop model is fed back to the crop dynamic irrigation water prediction model, the difference between W-W2 is solved, the Aquacrop model parameters are adjusted, the Aquacrop model is optimized, and it is iterated repeatedly until the difference is infinitely close to 0, thus achieving accurate prediction of the Aquacrop model. Dynamic irrigation water prediction model W d The specific expression is:

[0069] Where △ represents the slope of the water vapor pressure curve, in kPa / ℃; A represents the difference between the net radiation of the crop surface and the soil heat flux density, in MJ / (m 2 d); γ is the hygrometer constant, unit is kPa / ℃; w is the wind speed 2m above the ground, unit is m / s; v is the saturated pressure deficit, unit is kPa; T a Indicates the average temperature 2m above the ground, in °C; P f Indicates the future daily rainfall in mm, k c Represents the crop coefficient.

[0070] Step 5: Optimize the irrigation system with the goal of maximizing yield and minimizing irrigation volume. Using the Aquacrop model parameters when the W-W2 difference is infinitely close to 0 as calibration parameters, and maximizing crop yield and minimizing total irrigation water volume as the goal, optimize the irrigation decision-making mechanism for the actual irrigation area. With irrigation method, irrigation time, irrigation quota, and irrigation frequency as variables, the expression is: Where Y max Indicates the maximum crop yield, unit is t / hm 2 , W min Indicates the minimum total amount of irrigation water, unit is mm, D indicates the irrigation time, I indicates the single irrigation water volume, unit is mm, M indicates the irrigation method, including border irrigation, flood irrigation, drip irrigation, N indicates the number of irrigation times, I i represents the amount of irrigation water for the i-th time, and f(D,I,M,N) represents the function of yield.

[0071] Step 6: Smart irrigation decision-making system based on cloud-edge collaboration. Use the Raspberry Pi system and LoRa gateway as the MQTT protocol agent, place the data and algorithm model on the cloud, realize data collection and upload and control command issuance, and build a cloud-edge collaborative system for smart irrigation. Figure 4 As shown. The microservice architecture is as follows Figure 5 The following layers are shown:

[0072] Step 61: Production layer: Collect historical crop growth data, historical field management data, historical soil parameter data, historical yield, and historical and future meteorological data of the site.

[0073] Step 62: Data layer: collect and store production layer data based on the time series database.

[0074] Step 63: Application layer: Apply the data layer data to the algorithm model in steps 4 and 5 and the optimized irrigation decision mechanism, and support access to the data layer.

[0075] Step 64: Terminal layer: Decision makers control the valves through http: / / terminals or farmers manually control the valves to achieve multi-task collaboration, such as Figure 6 shown.

[0076] The present invention has the following technical effects:

[0077] 1. High model accuracy:

[0078] By coupling the crop dynamic irrigation water prediction model with the Aquacrop model and iterating repeatedly until the difference in the total irrigation water volume of the two models is infinitely close to 0, accurate prediction of the Aquacrop model can be achieved.

[0079] 2. Cloud-edge-device collaborative system with high efficiency and collaborative interaction:

[0080] Through multidisciplinary cross-integration, we have achieved breakthroughs in key technologies such as digital perception of farmland moisture information using integrated air-space-ground and air-ground technology, intelligent irrigation decision-making using hybrid human-machine technology, and precise control of cloud-edge-end collaborative irrigation systems. We have developed a smart farmland irrigation management service system that integrates edge computing and deep learning, and improved the precise and intelligent control capabilities of water and fertilizer. We have also achieved the organic integration of irrigation with agronomy and agricultural machinery technology, and improved soil fertility and grain production capacity.

[0081] In accordance with the above method, an embodiment of the present invention provides a crop irrigation system, comprising:

[0082] The acquisition module is used to obtain a sample set and future meteorological data of the site where the crop is located; the sample set includes historical growth data, historical field management data, historical soil parameter data, historical yield and historical meteorological data of the site where the crop is located.

[0083] The Aquacrop model training module is used to train the Aquacrop model using the sample set and the preset irrigation amount to obtain a trained Aquacrop model.

[0084] The first irrigation water predicted amount determination module is used to obtain the first irrigation water predicted amount of the crop based on the historical meteorological data of the site where the crop is located and the crop coefficient of the crop.

[0085] The dynamic irrigation water prediction model training module is used to take the future meteorological data of the site where the crop is located and the historical meteorological data of the site where the crop is located as input, and the first predicted amount of irrigation water as output, and use a deep learning algorithm to train the dynamic irrigation water prediction model to obtain a trained dynamic irrigation water prediction model.

[0086] The second irrigation water predicted amount determination module is used to input the future meteorological data of the site where the crop is located into the trained dynamic irrigation water prediction model to obtain the second irrigation water predicted amount.

[0087] The Aquacrop model optimization module is used to optimize the trained Aquacrop model parameters to obtain an optimized Aquacrop model with the difference between the preset irrigation amount of the crop and the second predicted irrigation water amount of the crop being 0 as the goal.

[0088] The irrigation module is used to irrigate with the optimized Aquacrop model with the goal of maximizing crop yield and minimizing the total amount of irrigation water.

[0089] In practical applications, the Aquacrop model training module specifically includes:

[0090] The Aquacrop model training unit is used to train the Aquacrop model using the preset irrigation amount and historical growth data, historical field management data, historical soil parameter data, and historical meteorological data of the site of the crops in the sample set as input, and using the historical yield of the crops in the sample set as output to obtain a trained Aquacrop model.

[0091] In practical applications, the first irrigation water prediction and determination module specifically includes:

[0092] The reference crop evapotranspiration calculation unit is used to obtain the reference crop evapotranspiration based on the historical meteorological data of the site where the crop is located and the Penman formula.

[0093] The first irrigation water prediction calculation unit is used to calculate the amount of irrigation water according to the formula W1=k c ET0-P calculates the first predicted amount of irrigation water for crops, where W1 represents the first predicted amount of irrigation water for crops, k c represents the crop coefficient of the crop, ET0 represents the evapotranspiration of the reference crop, and P represents the daily effective rainfall.

[0094] In practical applications, the calculation formula for the daily effective rainfall is:

[0095] Among them, P m It represents daily rainfall, and mm represents millimeters.

[0096] An embodiment of the present invention provides an electronic device, including:

[0097] A memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the crop irrigation method according to the above method embodiment.

[0098] An embodiment of the present invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the crop irrigation method as described in the above method embodiment is implemented.

[0099] At present, the integration of irrigation districts with modern information technology in terms of information monitoring and water use management is obviously insufficient. Therefore, the present invention uses cloud computing, edge computing, automatic control, electronic technology, etc. to integrate perception networks and cloud information to form an irrigation prescription map that matches the irrigation district. It combines deep learning and sensor perception technology to conduct real-time regulation of irrigation, and develops a cloud-edge collaborative intelligent irrigation service system to meet the collaborative operation needs of multiple irrigation tasks in the irrigation district. It realizes functions such as real-time irrigation forecast and real-time appropriate water use management, achieves "supply according to demand", and fully considers future rainfall to avoid irrigation waste caused by rain after irrigation and excessive irrigation as much as possible.

[0100] This paper constructs an Aquacrop optimization model based on deep learning and irrigation district hydrological models, forms irrigation prescription maps that match the irrigation district, and improves the irrigation decision-making mechanism for the actual irrigation district. Through cloud-edge-end collaboration, it achieves an efficient synchronization mechanism between irrigation district data perception, analysis, and decision-making, and builds a cloud-edge-end collaborative intelligent irrigation decision-making system with initial forecasting, early warning, rehearsal, and pre-planning capabilities.

[0101] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0102] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A crop irrigation method, characterized in that: include: Obtaining a sample set and future meteorological data for the site where the crop is located; the sample set includes historical growth data, historical field management data, historical soil parameter data, historical yield data, and historical meteorological data for the site where the crop is located; Using the sample set and the preset irrigation amount to train the Aquacrop model to obtain a trained Aquacrop model; Obtaining a first predicted amount of irrigation water for the crop based on historical meteorological data of the site where the crop is located and a crop coefficient of the crop; The future meteorological data of the site where the crop is located and the historical meteorological data of the site where the crop is located are used as input, and the first predicted amount of irrigation water is used as output, and a deep learning algorithm is used to train a dynamic irrigation water prediction model to obtain a trained dynamic irrigation water prediction model; the dynamic irrigation water prediction model W d The specific expression is: Among them, ET c Indicates crop water requirement, P f represents the future daily rainfall; Inputting future meteorological data of the site where the crop is located into the trained dynamic irrigation water prediction model to obtain a second irrigation water prediction amount; Taking the difference between the preset irrigation amount of the crop and the second predicted irrigation water amount of the crop as 0 as a target, optimizing the trained Aquacrop model parameters to obtain an optimized Aquacrop model; With the goal of maximizing crop yield and minimizing the total amount of irrigation water, the optimized Aquacrop model is used for irrigation. Specifically, the optimized Aquacrop model parameters are used as calibration parameters, and the irrigation decision mechanism is optimized with the goal of maximizing crop yield and minimizing the total amount of irrigation water. With irrigation method, irrigation time, irrigation quota, and irrigation times as variables, the expression is: Where Y max Indicates the maximum crop yield, W min Indicates the minimum total amount of irrigation water, D indicates the irrigation time, I indicates the amount of single irrigation water, M indicates the irrigation method, N indicates the number of irrigation times, I i represents the amount of irrigation water for the i-th time, and f(D,I,M,N) represents the function of yield.

2. The crop irrigation method according to claim 1, characterized in that: The Aquacrop model is trained using the sample set and the preset irrigation amount to obtain a trained Aquacrop model, specifically including: The Aquacrop model is trained using the preset irrigation amount and historical growth data, historical field management data, historical soil parameter data, and historical meteorological data of the site of the crops in the sample set as input, and using the historical yield of the crops in the sample set as output to obtain a trained Aquacrop model.

3. The crop irrigation method according to claim 1, characterized in that: Obtaining a first predicted amount of irrigation water for the crop according to historical meteorological data of the site where the crop is located and the crop coefficient of the crop specifically includes: The evapotranspiration of the reference crop is obtained based on the historical meteorological data of the site where the crop is located and the Penman formula; According to the formula W1=k c ET0-P calculates the first predicted amount of irrigation water for crops, where W1 represents the first predicted amount of irrigation water for crops, k c represents the crop coefficient of the crop, ET0 represents the evapotranspiration of the reference crop, and P represents the daily effective rainfall.

4. The crop irrigation method according to claim 3, characterized in that: The calculation formula for the daily effective rainfall is: Among them, P m It represents daily rainfall, and mm represents millimeters.

5. A crop irrigation system, characterized in that: include: An acquisition module is used to acquire a sample set and future meteorological data of the site where the crop is located; the sample set includes historical growth data, historical field management data, historical soil parameter data, historical yield data, and historical meteorological data of the site where the crop is located; An Aquacrop model training module is used to train the Aquacrop model using the sample set and the preset irrigation amount to obtain a trained Aquacrop model; A first irrigation water forecast amount determination module, configured to obtain a first irrigation water forecast amount for the crop based on historical meteorological data of the site where the crop is located and a crop coefficient of the crop; The dynamic irrigation water prediction model training module is used to use the future meteorological data of the site where the crop is located and the historical meteorological data of the site where the crop is located as input, and the first predicted amount of irrigation water as output, and to use a deep learning algorithm to train the dynamic irrigation water prediction model to obtain a trained dynamic irrigation water prediction model; the dynamic irrigation water prediction model W d The specific expression is: Among them, ET c Indicates crop water requirement, P f represents the future daily rainfall; A second irrigation water predicted amount determination module is configured to input future meteorological data of the site where the crop is located into the trained dynamic irrigation water prediction model to obtain a second irrigation water predicted amount; An Aquacrop model optimization module is used to optimize the trained Aquacrop model parameters to obtain an optimized Aquacrop model with the difference between the preset irrigation amount of the crop and the second predicted irrigation water amount of the crop being 0 as a target; The irrigation module is used to irrigate with the goal of maximizing crop yield and minimizing the total amount of irrigation water, using the optimized Aquacrop model. Specifically, the optimized Aquacrop model parameters are used as calibration parameters, and the irrigation method, irrigation time, irrigation quota, and irrigation frequency are used as variables to optimize the irrigation decision mechanism. The expression is: Where Y max Indicates the maximum crop yield, W min Indicates the minimum total amount of irrigation water, D indicates the irrigation time, I indicates the amount of single irrigation water, M indicates the irrigation method, N indicates the number of irrigation times, I i represents the amount of irrigation water for the i-th time, and f(D,I,M,N) represents the function of yield.

6. The crop irrigation system according to claim 5, characterized in that: The Aquacrop model training module specifically includes: The Aquacrop model training unit is used to train the Aquacrop model using the preset irrigation amount and historical growth data, historical field management data, historical soil parameter data, and historical meteorological data of the site of the crops in the sample set as input, and using the historical yield of the crops in the sample set as output to obtain a trained Aquacrop model.

7. The crop irrigation system according to claim 5, characterized in that: The first irrigation water pre-measurement determination module specifically includes: A reference crop evapotranspiration calculation unit, configured to obtain the reference crop evapotranspiration based on historical meteorological data of the site where the crop is located and the Penman formula; The first irrigation water prediction calculation unit is used to calculate the amount of irrigation water according to the formula W1=k c ET0-P calculates the first predicted amount of irrigation water for crops, where W1 represents the first predicted amount of irrigation water for crops, k c represents the crop coefficient of the crop, ET0 represents the evapotranspiration of the reference crop, and P represents the daily effective rainfall.

8. The crop irrigation system according to claim 7, characterized in that: The calculation formula for the daily effective rainfall is: Among them, P m It represents daily rainfall, and mm represents millimeters.

9. An electronic device, characterized in that: include: A memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the crop irrigation method according to any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that The device stores a computer program, which, when executed by a processor, implements the crop irrigation method according to any one of claims 1 to 4.

Images