Intelligent irrigation method and system based on soil moisture content prediction and storage medium

Abstract

The invention discloses an intelligent irrigation method and system based on soil moisture content prediction and a storage medium. The method comprises the following steps: collecting environment data related to farmland irrigation, wherein the environment data at least comprises soil humidity data reflecting a soil moisture state and crop information representing a crop growth state; based on the environmental data, constructing a historical data sequence of soil humidity changing along with time, and processing the historical data sequence by using a time sequence prediction model to obtain a soil humidity prediction value corresponding to at least one time point in a future set time period; comparing the soil humidity predicted value with the optimal water demand condition of the corresponding growth stage of the crop, and on the premise that the optimal water demand condition is met, performing optimization solution on irrigation parameters to generate irrigation decision information; and controlling irrigation equipment to execute variable irrigation operation according to the irrigation decision information.

Description

Technical Field

[0001] This application relates to the field of agricultural science and technology, and in particular to an intelligent irrigation method, system and storage medium based on soil moisture prediction. Background Technology

[0002] Water scarcity has become one of the key factors restricting the sustainable development of modern agriculture. Traditional irrigation methods have obvious limitations: On the one hand, timed irrigation supplies water at fixed intervals, failing to fully consider the actual water requirements of crops and changes in environmental conditions, easily leading to water waste or insufficient irrigation. On the other hand, irrigation methods based on real-time monitoring rely primarily on current or historical soil moisture data for irrigation decisions, lacking the ability to predict future water trends and exhibiting inherent irrigation lag. When soil moisture drops below the crop's water requirement threshold, the crop may have already experienced water stress for a certain period, thus affecting its growth and yield.

[0003] Existing research attempts to estimate soil moisture using meteorological information or simple models, but there are still some shortcomings, such as: limited prediction accuracy, making it difficult to fully reflect the changing patterns of soil moisture; a single irrigation decision-making method, making it difficult to balance water resource utilization efficiency and crop growth needs under different agricultural production conditions; and a disconnect between decision-making and actual irrigation implementation, failing to form an effective closed loop for crop water demand management.

[0004] Therefore, there is still an urgent need in this field for a technical solution that can improve existing irrigation methods to solve the problems of irrigation delays, low water use efficiency, and the inability to accurately meet crop water requirements. Summary of the Invention

[0005] To address the aforementioned technical problems, this application provides an intelligent irrigation method, system, and storage medium based on soil moisture prediction.

[0006] The technical solution provided in this application is described below: The first aspect of this application provides a smart irrigation method based on soil moisture prediction, the method comprising: Collect environmental data related to farmland irrigation, including at least soil moisture data reflecting soil moisture status and crop information characterizing crop growth status; Based on the environmental data, a historical data sequence of soil moisture changes over time is constructed, and the historical data sequence is processed using a time series prediction model to obtain a predicted value of soil moisture at least one point in time within a future set period. The predicted soil moisture value is compared with the optimal water requirement conditions for the corresponding growth stage of the crop, and the irrigation parameters are optimized and solved under the premise of meeting the optimal water requirement conditions to generate irrigation decision information. Based on the irrigation decision information, the irrigation equipment is controlled to perform variable irrigation operations.

[0007] Optionally, the time series prediction model includes a prediction model based on a recurrent neural network structure, used to characterize the long-term dependence of soil moisture on changes over time.

[0008] Optionally, the step of optimizing the irrigation parameters to generate irrigation decision information includes: Based on the predicted soil moisture values, an objective function is constructed that includes crop water stress level, irrigation water consumption, and irrigation energy consumption. Under the constraints of the objective function, the irrigation parameters are optimized through multi-objective solutions to generate irrigation decision information that satisfies the optimal water demand conditions.

[0009] Optionally, the collection of environmental data related to farmland irrigation includes: Acquire images of the crop canopy; Image analysis is performed on crop canopy images to determine the current growth status of the crop and generate corresponding growth stage information.

[0010] Optionally, the training process of the soil moisture prediction model includes the following sub-steps: The features of each time step in the input sequence formed by historical environmental data are analyzed, and different weights are assigned to different time step features through an attention mechanism to highlight key time step features that have a significant impact on soil moisture changes. The weighted input sequence is fed into a recurrent neural network model, which is then used to learn the long-term dependencies in the input sequence. Based on the learning results of the recurrent neural network, a soil moisture prediction sequence corresponding to multiple time points within a future set time period is output.

[0011] Optionally, the multi-objective optimization solution employs the non-dominated sorting genetic algorithm NSGA-II.

[0012] A second aspect of this application provides an intelligent irrigation system based on soil moisture prediction, the system comprising: The data acquisition unit is used to collect environmental data related to farmland irrigation. The environmental data includes at least soil moisture data reflecting the soil moisture status and crop information characterizing the crop growth status. The model prediction unit constructs a historical data sequence of soil moisture changes over time based on the environmental data, and processes the historical data sequence using a time series prediction model to obtain a predicted value of soil moisture at least one time point within a future set period. The multi-objective solution unit is used to compare the predicted soil moisture value with the optimal water requirement conditions for the corresponding growth stage of the crop, and optimize the irrigation parameters under the premise of meeting the optimal water requirement conditions to generate irrigation decision information. The control unit is used to control the irrigation equipment to perform variable irrigation operations based on the irrigation decision information.

[0013] Optionally, the multi-objective solving unit is specifically used for: Based on the predicted soil moisture values, an objective function is constructed that includes crop water stress level, irrigation water consumption, and irrigation energy consumption. Under the constraints of the objective function, the irrigation parameters are optimized through multi-objective solutions to generate irrigation decision information that satisfies the optimal water demand conditions.

[0014] A third aspect of this application provides an intelligent irrigation system based on soil moisture prediction, the system comprising: Processor, memory, input / output units, and bus; The processor is connected to the memory, the input / output unit, and the bus; The memory stores a program, which the processor invokes to execute the first aspect and any one of the optional methods in the first aspect.

[0015] A fourth aspect of this application provides a computer-readable storage medium on which a program is stored, which, when executed on a computer, performs the methods of the first aspect and any one of the first aspects.

[0016] As can be seen from the above technical solutions, this application has the following beneficial effects: 1. By analyzing historical environmental data and crop growth status, it is possible to predict future soil moisture trends, so that irrigation decisions no longer rely on single real-time monitoring data, thereby overcoming the lag problem of traditional irrigation methods and enabling timely adjustment of irrigation operations before crops suffer water stress.

[0017] 2. By comprehensively considering multiple objectives such as crop water requirements, irrigation water consumption, and energy consumption, irrigation strategies are generated to save water resources and reduce irrigation energy consumption while ensuring crop growth, thereby improving the overall efficiency of irrigation operations.

[0018] 3. The method of this application can dynamically adjust the irrigation strategy according to the water requirements of crops at different growth stages, so that the water supply is more in line with the actual needs of crops, improve crop growth conditions, and ensure yield and quality. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a schematic flowchart of an embodiment of the intelligent irrigation method based on soil moisture prediction provided in this application; Figure 2 This is a schematic flowchart of an embodiment of step S102 in the intelligent irrigation method based on soil moisture prediction provided in this application; Figure 3 This is a schematic flowchart of an embodiment of step S103 in the intelligent irrigation method based on soil moisture prediction provided in this application; Figure 4 This is a schematic diagram of an embodiment of the intelligent irrigation system based on soil moisture prediction provided in this application; Figure 5 This is a schematic diagram of another embodiment of an intelligent irrigation system based on soil moisture prediction provided in this application. Detailed Implementation

[0021] In this embodiment, the intelligent irrigation method can be implemented by various different entities. These entities can be individual irrigation control systems, computers, embedded processing devices, edge computing devices, or servers, or they can be a system composed of the aforementioned devices and irrigation equipment. The entity can include hardware modules, software modules, or a combination of both, executing the various steps of the method of the present invention through software instructions or hardware control logic.

[0022] In practical applications, each step of the method can be completed entirely by a single executing entity or collaboratively by multiple executing entities, without limitation on specific implementation methods or specific equipment types. Through this unrestricted executing entity design, the method of the present invention can adapt to farmland of different sizes, different types of irrigation equipment, and different information processing platforms, ensuring the flexibility and scalability of the method.

[0023] Please see Figure 1 This application first provides an embodiment of an intelligent irrigation method based on soil moisture prediction, which includes: S101. Collect environmental data related to farmland irrigation, wherein the environmental data includes at least soil moisture data reflecting soil moisture status and crop information characterizing crop growth status. In this embodiment, the implementing entity first deploys environmental sensors and data acquisition devices in the farmland area to collect environmental data related to irrigation. The environmental data includes at least: Soil moisture data: Soil moisture content information is obtained by using soil moisture sensors at different depths, such as moisture data for the surface layer (0–10 cm), middle layer (10–30 cm), and deep layer (30–50 cm). The sensor can sample data at fixed time intervals (such as once every 10 minutes) to form a time series.

[0024] Crop Information: Indicators such as crop type, growth stage, and leaf area index can be recorded manually or collected by devices such as sensors and cameras. In this embodiment, crop information can be obtained by acquiring crop canopy images and performing image analysis to determine the crop growth stage, such as whether the crop is in the seedling stage, flowering stage, or fruiting stage.

[0025] Other environmental data: Meteorological data, such as air temperature and humidity, light intensity, and rainfall; Soil electrical conductivity, temperature, and other parameters are used to further aid in understanding soil moisture.

[0026] The collected data can be centrally managed through wireless transmission or local acquisition devices, and can be stored in a database or local cache to provide a data foundation for subsequent prediction steps.

[0027] In an optional embodiment, the step includes acquiring images of the crop canopy; performing image analysis on the crop canopy images to determine the current growth status of the crop and generating corresponding growth stage information.

[0028] In this optional embodiment, crop phenotypic information is obtained by acquiring images of the crop canopy. The acquired images can come from fixed cameras, drone aerial photography, or mobile imaging equipment, and can cover different growth areas and heights to ensure comprehensive observation of the crop canopy.

[0029] The acquired canopy images undergo image analysis and processing, including image denoising, color segmentation, texture recognition, and leaf area extraction, to extract feature information related to crop growth. For example, leaf color, coverage, and density can be analyzed to assess crop growth stage characteristics.

[0030] Based on image analysis, crop growth is divided into different stages, such as seedling stage, vegetative growth stage, flowering stage, and fruiting stage, and corresponding growth stage information is generated. This generated growth stage information can be combined with historical growth records for subsequent soil moisture prediction and irrigation decisions, ensuring that irrigation strategies better meet the actual water requirements of the crops.

[0031] For example, in a certain farmland area, by analyzing crop canopy images, it can be identified that the crop has high leaf coverage and early flower buds, indicating that the crop is in the flowering stage. This growth stage information will be used as input parameters for soil moisture prediction and irrigation decision generation, ensuring that irrigation amounts meet the crop's high water demand period.

[0032] S102. Based on the environmental data, construct a historical data sequence of soil moisture changes over time, and use a time series prediction model to process the historical data sequence to obtain a predicted value of soil moisture at least one time point within a future set period. In this step, after obtaining environmental data, the implementing entity organizes the collected historical data into a multidimensional time series and preprocesses the data, including missing value imputation, outlier removal, and standardization, to ensure data quality. The processed data is then input into a time series prediction model to predict soil moisture changes over a specified future period, thereby obtaining predicted soil moisture values ​​for at least one future time point. The prediction results can generate a soil moisture change curve, providing a basis for determining whether crops may be subject to water stress.

[0033] For example, if the current soil moisture is 25%, and the prediction model outputs 24%, 22%, 21%, 20%, 18%, and 17% for the next 6 hours, it can be used to determine whether the irrigation amount needs to be adjusted in advance.

[0034] In an optional embodiment, the time series prediction model includes a prediction model based on a recurrent neural network structure for characterizing the long-term dependence of soil moisture on changes over time.

[0035] In this optional embodiment, the time series prediction model employs a prediction model based on a recurrent neural network (RNN) structure to characterize the long-term dependence of soil moisture changes over time. This model can learn from historical environmental data sequences, capture the dynamic changes in soil moisture over time, and generate a predicted soil moisture sequence for a predetermined future period to assess the future water demand risk of crops.

[0036] During implementation, collected multi-depth soil moisture data, crop growth stage information, and relevant meteorological data were compiled into a multi-dimensional time series and preprocessed, including missing value imputation, outlier removal, and standardization. The processed data was then input into a recurrent neural network model. Through its hidden layer state propagation mechanism, the model can identify dependencies between different time steps, thereby predicting future soil moisture changes. The prediction results can output soil moisture values ​​for multiple future time points and form soil moisture change curves, providing a reference for subsequent irrigation decisions.

[0037] For example, suppose in a certain farmland area, the current soil moisture is 28% for the topsoil, 26% for the middle layer, and 24% for the deep layer, and the crop is in the flowering stage. After inputting this data into a prediction model based on a recurrent neural network, a predicted sequence of soil moisture for each hour over the next 6 hours can be obtained. For example, the predicted values ​​for the topsoil are 27%, 25%, 24%, 22%, 21%, and 20%, with the predicted values ​​for the middle and deep layers output simultaneously. Based on the prediction sequence, it can be determined that the soil moisture may fall below the crop's water requirement threshold within the next 2–3 hours, thus providing a basis for generating irrigation strategies.

[0038] In this embodiment, the recurrent neural network structure can be a single-layer or multi-layer RNN, and it can also be used in conjunction with other time series processing modules to improve prediction accuracy and stability. At the same time, this model can flexibly adapt to different crop types, soil conditions, and environmental data inputs, ensuring the reliability and usability of the prediction results.

[0039] See Figure 2 In a more specific embodiment, the training process of the soil moisture prediction model includes the following sub-steps: S1021. Analyze the features of each time step in the input sequence formed by historical environmental data, and assign different weights to different time step features through an attention mechanism to highlight key time step features that have a significant impact on soil moisture changes. The features of each time step in the input sequence formed by historical environmental data are analyzed to identify key time steps that have a significant impact on soil moisture changes. For example, recent rainfall, temperature changes, or crop irrigation records may have a significant impact on future soil moisture. By assigning different weights to the features of each time step through an attention mechanism, the model can highlight the key factors characterizing soil moisture changes and improve prediction accuracy.

[0040] S1022. Input the weighted input sequence into the recurrent neural network model, and use the recurrent neural network to learn the long-term dependencies in the input sequence; The attention-weighted input sequence is fed into a recurrent neural network (RNN) model. Through its hidden state propagation mechanism, the RNN effectively learns the long-term dependencies between different time steps in the input sequence, capturing the dynamic changes in soil moisture over time. During training, historical soil moisture data and environmental characteristics can be used to optimize the network parameters, enabling the model to accurately predict future soil moisture trends.

[0041] S1023. Based on the learning results of the recurrent neural network, output the soil moisture prediction sequence corresponding to multiple time points within a future set time period.

[0042] Based on the learning results of a recurrent neural network, a predicted sequence of soil moisture at multiple time points within a specified future period is output. This predicted sequence forms a curve showing future soil moisture changes, providing a reference for irrigation decisions. The prediction results can be used to determine the crop's water supply and demand in the future, thereby guiding the adjustment of subsequent irrigation parameters.

[0043] S103. The predicted soil moisture value is compared with the optimal water requirement conditions for the corresponding growth stage of the crop, and the irrigation parameters are optimized and solved under the premise of meeting the optimal water requirement conditions to generate irrigation decision information. The predicted soil moisture value is compared with the optimal water requirement conditions corresponding to the current growth stage of the crop. If the predicted value indicates that the soil moisture may be lower than the lower limit of crop water requirement, the irrigation parameters are optimized to generate irrigation decision information. Irrigation parameters include irrigation time, irrigation amount, and possible farmland zoning strategies, and the irrigation amount can be flexibly adjusted according to different soil moisture ranges. During the comparison and solution process, soil moisture change trends, crop water requirement characteristics, and historical irrigation information can be comprehensively considered to ensure the rationality and accuracy of the irrigation strategy.

[0044] For example, if a forecast shows that soil moisture will drop to 17% in the next 6 hours and the crop's lower water requirement is 20%, an irrigation decision can be generated: start irrigation in the next hour, with an irrigation volume of 500 liters per acre.

[0045] See Figure 3 In an optional embodiment, step S103 is implemented by including: S1031. Based on the predicted soil moisture value, construct an objective function that includes crop water stress degree, irrigation water consumption and irrigation energy consumption. In this step, a multi-objective evaluation system reflecting crop water demand is established based on the predicted soil moisture values. The objective function may include at least three aspects: Crop water stress level is used to quantify the potential impact of future soil moisture deficiency on crop growth. Irrigation water consumption is used to assess the efficiency of irrigation resource utilization. Irrigation energy consumption is used to measure the energy consumed by irrigation equipment during operation.

[0046] By unifying these objectives, a comprehensive objective function for evaluating the merits of irrigation strategies can be formed, which can then be used to optimize irrigation parameters.

[0047] S1032, and under the constraints of the objective function, perform multi-objective optimization of the irrigation parameters to generate irrigation decision information that satisfies the optimal water demand conditions.

[0048] Under the constraints of the objective function and the requirement to meet the optimal water requirements of crops, irrigation parameters are optimized to generate final irrigation decision information. Irrigation parameters include irrigation start time, irrigation volume, and water allocation strategies for different areas of farmland. During the optimization process, future soil moisture trends, crop water requirements, and historical irrigation records can be comprehensively considered to ensure that the generated irrigation strategy meets crop growth needs while also taking into account water resource and energy utilization efficiency.

[0049] Assuming that the predicted soil moisture in the next 6 hours will be lower than the crop water requirement threshold, an objective function is constructed to comprehensively evaluate crop water stress, irrigation water consumption, and energy consumption. After multi-objective optimization, the irrigation decision is generated as follows: start irrigation in the next 1 hour, irrigate 500 liters of water per acre, and adjust the irrigation amount in different areas according to the soil moisture distribution to balance crop water requirement and resource utilization efficiency.

[0050] In this embodiment, the multi-objective optimization solution can combine historical environmental data and crop growth stage information to dynamically adjust the irrigation strategy, realize forward-looking irrigation control, and provide accurate decision information for the actual execution of irrigation equipment.

[0051] Specifically, in one optional embodiment, the multi-objective optimization solution employs the Non-Dominated Sorting Genetic Algorithm (NSGA-II). This algorithm encodes irrigation parameters (including irrigation time, irrigation amount, and water allocation in farmland zones) and simultaneously optimizes multiple objectives under the constraints of the objective function, such as crop water stress, irrigation water consumption, and irrigation energy consumption.

[0052] During the optimization process, NSGA-II selects the Pareto front solution set through non-dominated sorting and crowding distance calculation, generating irrigation strategies that meet the optimal water requirements of crops. This algorithm can balance crop water requirements with water resource and energy utilization efficiency, achieving balanced optimization of irrigation strategies.

[0053] In the foregoing embodiments, the specific implementation process of the method provided in this application has been described in detail. It should be understood that, to implement the above method, this application also provides a corresponding system and a computer-readable storage medium, the technical concept of which is consistent with the foregoing method embodiments, both used to implement all or part of the steps in the foregoing method. The technical solutions of the system and computer-readable storage medium involved in this application will be further described below in conjunction with specific embodiments.

[0054] S104. Based on the irrigation decision information, control the irrigation equipment to perform variable irrigation operations.

[0055] Based on the generated irrigation decision information, the irrigation equipment is controlled to complete the irrigation operation. During the irrigation process, the water volume can be adjusted according to different soil moisture ranges to achieve variable irrigation. After the irrigation operation is completed, soil moisture and environmental data can continue to be collected to provide a reference for subsequent prediction and decision-making, enabling dynamic adjustments.

[0056] See Figure 4 This application provides an intelligent irrigation system based on soil moisture prediction, the system comprising: The data acquisition unit 401 is used to collect environmental data related to farmland irrigation. The environmental data includes at least soil moisture data reflecting the soil moisture status and crop information characterizing the crop growth status. The model prediction unit 402 constructs a historical data sequence of soil moisture changes over time based on the environmental data, and processes the historical data sequence using a time series prediction model to obtain a predicted value of soil moisture at least one time point within a future set period. The multi-objective solution unit 403 is used to compare the predicted soil moisture value with the optimal water requirement conditions for the corresponding growth stage of the crop, and optimize the irrigation parameters under the premise of meeting the optimal water requirement conditions to generate irrigation decision information. The control unit 404 is used to control the irrigation equipment to perform variable irrigation operations based on the irrigation decision information.

[0057] Optionally, the time series prediction model includes a prediction model based on a recurrent neural network structure, used to characterize the long-term dependence of soil moisture on changes over time.

[0058] Optionally, the multi-objective solving element 403 is specifically used for Based on the predicted soil moisture values, an objective function is constructed that includes crop water stress level, irrigation water consumption, and irrigation energy consumption. Under the constraints of the objective function, the irrigation parameters are optimized through multi-objective solutions to generate irrigation decision information that satisfies the optimal water demand conditions.

[0059] Optionally, the data acquisition unit 401 is specifically used for: Acquire images of the crop canopy; Image analysis is performed on crop canopy images to determine the current growth status of the crop and generate corresponding growth stage information.

[0060] Optionally, the training process of the soil moisture prediction model includes the following sub-steps: The features of each time step in the input sequence formed by historical environmental data are analyzed, and different weights are assigned to different time step features through an attention mechanism to highlight key time step features that have a significant impact on soil moisture changes. The weighted input sequence is fed into a recurrent neural network model, which is then used to learn the long-term dependencies in the input sequence. Based on the learning results of the recurrent neural network, a soil moisture prediction sequence corresponding to multiple time points within a future set time period is output.

[0061] Optionally, the multi-objective optimization solution employs the non-dominated sorting genetic algorithm NSGA-II.

[0062] Please see Figure 5 This application also provides an intelligent irrigation system based on soil moisture prediction, comprising: Processor 501, memory 502, input / output unit 503, bus 504; The processor 501 is connected to the memory 502, the input / output unit 503, and the bus 504; The memory 502 stores a program, and the processor 501 calls the program to execute any of the methods described above.

[0063] This application also relates to a computer-readable storage medium on which a program is stored, which, when run on a computer, causes the computer to perform any of the methods described above.

[0064] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0065] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0066] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0067] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0068] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A smart irrigation method based on soil moisture prediction, characterized in that, The method includes: Collect environmental data related to farmland irrigation, including at least soil moisture data reflecting soil moisture status and crop information characterizing crop growth status; Based on the environmental data, a historical data sequence of soil moisture changes over time is constructed, and the historical data sequence is processed using a time series prediction model to obtain a predicted value of soil moisture at least one point in time within a future set period. The predicted soil moisture value is compared with the optimal water requirement conditions for the corresponding growth stage of the crop, and the irrigation parameters are optimized and solved under the premise of meeting the optimal water requirement conditions to generate irrigation decision information. Based on the irrigation decision information, the irrigation equipment is controlled to perform variable irrigation operations.

2. The intelligent irrigation method based on soil moisture prediction as described in claim 1, characterized in that, The time series prediction model includes a prediction model based on a recurrent neural network structure, used to characterize the long-term dependence of soil moisture on changes over time.

3. The intelligent irrigation method based on soil moisture prediction as described in claim 1, characterized in that, The process of optimizing irrigation parameters to generate irrigation decision information includes: Based on the predicted soil moisture values, an objective function is constructed that includes crop water stress level, irrigation water consumption, and irrigation energy consumption. Under the constraints of the objective function, the irrigation parameters are optimized through multi-objective solutions to generate irrigation decision information that satisfies the optimal water demand conditions.

4. The intelligent irrigation method based on soil moisture prediction according to claim 1, characterized in that, The collected environmental data related to farmland irrigation includes: Acquire images of the crop canopy; Image analysis is performed on the canopy images to determine the current growth status of the crop and generate corresponding growth stage information.

5. The intelligent irrigation method based on soil moisture prediction according to claim 2, characterized in that, The training process of the soil moisture prediction model includes the following sub-steps: The features of each time step in the input sequence formed by historical environmental data are analyzed, and different weights are assigned to different time step features through an attention mechanism to highlight key time step features that have a significant impact on soil moisture changes. The weighted input sequence is fed into a recurrent neural network model, which is then used to learn the long-term dependencies in the input sequence. Based on the learning results of the recurrent neural network, a soil moisture prediction sequence corresponding to multiple time points within a future set time period is output.

6. The intelligent irrigation method based on soil moisture prediction according to claim 3, characterized in that, The multi-objective optimization solution uses the non-dominated sorting genetic algorithm NSGA-II.

7. A smart irrigation system based on soil moisture prediction, characterized in that, The system includes: The data acquisition unit is used to collect environmental data related to farmland irrigation. The environmental data includes at least soil moisture data reflecting the soil moisture status and crop information characterizing the crop growth status. The model prediction unit constructs a historical data sequence of soil moisture changes over time based on the environmental data, and processes the historical data sequence using a time series prediction model to obtain a predicted value of soil moisture at least one time point within a future set period. The multi-objective solution unit is used to compare the predicted soil moisture value with the optimal water requirement conditions for the corresponding growth stage of the crop, and optimize the irrigation parameters under the premise of meeting the optimal water requirement conditions to generate irrigation decision information. The control unit is used to control the irrigation equipment to perform variable irrigation operations based on the irrigation decision information.

8. The intelligent irrigation method based on soil moisture prediction according to claim 3, characterized in that, The multi-objective solution unit is specifically used for: Based on the predicted soil moisture values, an objective function is constructed that includes crop water stress level, irrigation water consumption, and irrigation energy consumption. Under the constraints of the objective function, the irrigation parameters are optimized through multi-objective solutions to generate irrigation decision information that satisfies the optimal water demand conditions.

9. A smart irrigation system based on soil moisture prediction, characterized in that, The device includes: Processor, memory, input / output units, and bus; The processor is connected to the memory, the input / output unit, and the bus; The memory stores a program, which the processor invokes to perform the method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium contains a program that, when executed on a computer, performs the method as described in any one of claims 1 to 6.

Images