An intelligent crop water consumption prediction method for irrigation areas

Abstract

This invention discloses an intelligent method for predicting crop water consumption in irrigation areas, belonging to the field of crop water consumption prediction technology. The method includes data acquisition, data processing, crop water consumption estimation, and crop water consumption prediction model construction. Data acquisition involves obtaining meteorological data, soil data, vegetation index, and crop data for the irrigation area. Interpolation and resampling methods are used to unify the spatial resolution of all data and align them temporally. The SEBAL model is used to estimate daily evapotranspiration, i.e., crop water consumption. A Transformer model is used to model crop water consumption, meteorological data, soil data, and crop growth information. Through these steps, this invention can achieve high-precision and rapid prediction of the future water consumption of different crops by inputting planting structure, planting time, soil data, and meteorological forecast data, providing strong technical support for water conservation and precision agricultural irrigation in irrigation areas.

Description

Technical Field

[0001] This invention relates to the field of crop water consumption prediction technology, and in particular to an intelligent method for predicting crop water consumption in irrigation districts. Background Technology

[0002] Crop water consumption forecasting is a core component of agricultural water resource management. By integrating multi-source data such as meteorological, soil, and crop growth data, accurate predictions of crop water requirements can effectively achieve water conservation in farmland, improve water resource utilization efficiency, and ensure sustainable agricultural development. Current research on crop water consumption primarily relies on historical data modeling to estimate past water consumption, resulting in a severe lack of timeliness in setting irrigation water usage amounts and an inability to adapt to dynamic agricultural production needs.

[0003] Existing methods for estimating crop water consumption, such as the crop coefficient method, water balance method, and energy balance method, have significant technical shortcomings: some methods do not consider the inherent differences in evapotranspiration among different crops, resulting in estimation results that lack crop specificity; while some methods use empirical coefficients to optimize the estimation of water consumption for different crops, they ignore the differences in evapotranspiration of the same crop under different soil conditions, ultimately leading to a significant reduction in the accuracy of crop water consumption estimation, making it difficult to meet the actual needs of precision irrigation in irrigation districts.

[0004] Given that irrigation districts are important grain-producing areas in the country, and also face the common problem of water scarcity and the urgent need for accurate water consumption prediction in order to rationally allocate irrigation water resources, the development of a high-precision, high-timeliness crop water consumption prediction method that is adapted to the complex growing environment of irrigation districts has become an urgent need in the field of agricultural water-saving technology. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent method for predicting crop water consumption in irrigation areas, which solves the problems of poor timeliness, low estimation accuracy, and failure to comprehensively consider the differences between crops and soil environment in existing methods, and achieves high-precision and rapid prediction of the future water consumption of different crops in irrigation areas.

[0006] To achieve the above objectives, this invention provides an intelligent method for predicting crop water consumption in irrigation areas, comprising the following steps: S1: Data collection, through a combination of data acquisition by agricultural management departments and remote sensing inversion, collects multi-source daily-scale data on crop water consumption in the irrigation area, including meteorological data, soil data, vegetation index and crop data; S2: Data processing, standardizing the collected multi-source data to achieve uniformity of data spatial resolution and precise alignment of time and spatial dimensions; S3: Crop water consumption estimation. Based on the processed standardized data, the daily evapotranspiration of crops is calculated using the principle of surface energy balance and the SEBAL model, which is used as the actual water consumption of crops. S4: Crop water consumption prediction model construction. A Transformer model is built as the core prediction model. Through dataset construction, model training and accuracy verification, a usable crop water consumption prediction model is obtained. Inputting planting structure, planting time, soil data and weather forecast data can output the prediction results of future crop water consumption.

[0007] Preferably, step S1 specifically includes: S11: Meteorological data collection, collecting daily-scale observation data from meteorological stations within the irrigation area, including rainfall, surface temperature, air temperature, wind speed, and relative humidity; S12: Soil data collection, collecting soil data within the irrigation area, including soil salinity, soil type and soil texture; S13: Vegetation index extraction. The daily-scale vegetation indices of the irrigation area were extracted using Landsat 8 satellite sensors, including normalized vegetation index, enhanced vegetation index, green index, normalized water index, and soil-adjusted vegetation index. S14: Crop data collection, collecting crop data within the irrigation area, including crop planting time, growth and development time nodes, and planting structure data.

[0008] Preferably, step S2 specifically includes: S21: Interpolation and resampling methods are used to unify all data to a spatial resolution of 30 meters; S22: Align all data in time, organize crop planting time into raster data with a spatial resolution of 30 meters, and spatially align it with planting structure data.

[0009] Preferably, step S3 specifically includes: S31: Using remote sensing data and meteorological data, calculate net radiation, surface heat flux and sensible heat flux based on the principle of surface energy balance; S32: Calculate diurnal evapotranspiration based on the energy balance equation of the SEBAL model, using the following formula:

[0010] in, Net radiation, For surface heat flux, For sensible heat flux, Evaporation rate It is the difference between the aerodynamic surface temperature and the reference altitude temperature. Surface temperature, and This is an empirical coefficient. It is air density. It is the specific heat of air. It is the aerodynamic drag that accounts for heat transfer between the aerodynamic surface and the reference altitude. For surface albedo, and For incident and outgoing longwave radiation, It is the surface thermal emissivity. It is the actual direct and diffuse solar radiation flux reaching the Earth's surface.

[0011] Preferably, the construction of the crop water consumption prediction model in step S4 specifically includes the following sub-steps: S41: Construct a dataset, using the estimated crop water consumption as label data and meteorological data, soil data, and crop growth information as feature data, and divide them into training set, validation set, and test set. S42: Training the model. The Transformer model is trained using the training set, and the model parameters are adjusted in real time using the validation set to optimize the model's fitting effect. S43: Accuracy verification. Input the feature data from the test set into the trained Transformer model, output the predicted value of crop water consumption, calculate the error between the predicted value and the actual water consumption in the test set, verify the prediction accuracy of the model, and form the final crop water consumption prediction model.

[0012] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects: (1) Improve the timeliness and foresight of forecasts: break through the limitations of existing technologies that estimate past water consumption based on historical data, and achieve the prediction of future crop water consumption by combining forecast models with meteorological forecast data, providing advance notice for irrigation planning in irrigation districts and improving the timeliness of water resource allocation.

[0013] (2) Improve estimation and prediction accuracy: Take into account the differences in evapotranspiration of different crops and the differences in evapotranspiration of the same crop under different soil conditions. At the same time, integrate multi-source data such as meteorology, vegetation, and crop growth to ensure accuracy throughout the entire process from data collection to model construction, and solve the problem of low accuracy of existing methods.

[0014] (3) Enhance model adaptability: The Transformer model is used to model crop water consumption and multiple influencing factors. This model can effectively capture the complex relationship between multiple factors and has strong adaptability to climate change, soil environment differences and water demand characteristics of different growth and development stages of crops, and is suitable for the complex agricultural production environment of irrigation areas.

[0015] (4) Achieve rapid and accurate prediction: Through standardized data processing and efficient model architecture, it is possible to rapidly predict crop water consumption in irrigation areas, and the prediction results are highly accurate. This can provide strong technical support for water-saving irrigation and precision agricultural irrigation in irrigation areas, and help optimize the allocation and sustainable use of agricultural water resources.

[0016] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of an intelligent irrigation area crop water consumption prediction method according to an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] Example like Figure 1 As shown, this embodiment discloses an intelligent method for predicting crop water consumption in irrigation areas. This method is applicable to the prediction of water consumption of field crops in various agricultural irrigation areas. It can accurately predict the water consumption of single or multiple crops in the irrigation area, providing technical support for the formulation of precision irrigation plans and the optimal allocation of agricultural water resources. This embodiment is implemented in sequence according to the steps of data collection, data processing, crop water consumption estimation, and crop water consumption prediction model construction. The specific implementation methods of each step are as follows [A1]: S1: Data Acquisition. This involves collecting multi-source diurnal data on crop water consumption within the irrigation district through a combination of data acquisition from agricultural management departments and remote sensing inversion. Data includes meteorological data, soil data, vegetation indices, and crop data, ensuring data coverage of the entire irrigation district and alignment with the crop growth cycle. This process is divided into the following sub-steps: S11: Collect daily-scale observation data from meteorological stations within the irrigation area, including rainfall, surface temperature, air temperature, wind speed, and relative humidity. The data source is the local meteorological management department to ensure the temporal continuity and spatial coverage of the meteorological data.

[0022] S12: By combining on-site soil sampling and testing with retrieval of soil databases from local agricultural management departments, soil data for the entire irrigation area is collected, including soil salinity, soil type, and soil texture data, to achieve grid-based distribution of soil data within the irrigation area.

[0023] S13: Acquire remote sensing images of the irrigation area taken by the Landsat 8 satellite sensor, interpret and process the remote sensing images using professional remote sensing data processing software, and extract the daily-scale normalized vegetation index (NDVI), enhanced vegetation index (EVI), green index (GI), normalized water index (NDWI), and soil-adjusted vegetation index (SAVI) of the irrigation area. Step S14: Obtain basic crop data related to the irrigation area from the local agricultural and rural management department, including crop planting time, time nodes of each growth and development stage, and planting structure data of the irrigation area, to clarify the crop types and planting layout of different plots.

[0024] S2: Data processing, which standardizes the collected multi-source data to eliminate differences in spatial resolution and temporal dimensions, achieving uniform spatial resolution and precise alignment of temporal and spatial dimensions; specifically including: S21: For different types of raw data, appropriate interpolation and resampling methods are used for processing. Spatial interpolation is performed on meteorological and soil non-raster data, and resampling is performed on remote sensing vegetation index raster data. All data are unified to a spatial resolution of 30 meters to form 30-meter × 30-meter gridded data of the entire irrigation area.

[0025] S22: Organize the timeline of all gridded data that have achieved unified spatial resolution, unify the time scale of various types of data to the daily scale, and achieve time alignment of all data; at the same time, convert crop planting time data into 30-meter spatial resolution raster data, and accurately align it with the planting structure raster data in spatial dimension according to the layout of irrigation area plots, so as to ensure that the meteorological, soil, vegetation and crop data in the same raster unit correspond one-to-one.

[0026] S3: Crop water consumption estimation. Based on processed standardized data, and combining the principles of land surface energy balance and the SEBAL model, daily crop evapotranspiration is calculated as the actual crop water consumption; specifically including: S31: Substitute the gridded remote sensing vegetation index data and meteorological data of the irrigation area into the surface energy balance calculation model, and calculate the daily net radiation of each grid cell in the irrigation area based on the principle of surface energy balance; then, combine the surface reflectivity and solar radiation related information of each grid cell to further calculate the corresponding surface heat flux and sensible heat flux.

[0027] S32: Substitute the net radiation, surface heat flux, and sensible heat flux calculated in step S31 into the SEBAL model. Using the model's built-in energy balance equations, calculate the daily evapotranspiration of each grid cell in the irrigation district. This evapotranspiration represents the actual daily water consumption of the corresponding crop plot, thus completing the full estimation of crop water consumption across the entire irrigation district. The formula for calculating daily evapotranspiration is:

[0028] in, Net radiation, For surface heat flux, For sensible heat flux, Evaporation rate It is the difference between the aerodynamic surface temperature and the reference altitude temperature. Surface temperature, and This is an empirical coefficient. It is air density. It is the specific heat of air. It is the aerodynamic drag that accounts for heat transfer between the aerodynamic surface and the reference altitude. For surface albedo, and For incident and outgoing longwave radiation, It is the surface thermal emissivity. It is the actual direct and diffuse solar radiation flux reaching the Earth's surface.

[0029] S4: Crop water consumption prediction model construction. A Transformer model is built as the core prediction model. Through dataset construction, model training and accuracy verification, a usable crop water consumption prediction model is obtained. Inputting planting structure, planting time, soil data and weather forecast data can output the prediction results of future crop water consumption.

[0030] The construction of a crop water consumption prediction model includes the following sub-steps: S41: Construct a dataset, using the crop water consumption data estimated in step S3 as label data, and meteorological data, soil data, and crop growth information as feature data. Randomly divide all the label data and feature data into training set, validation set, and test set according to a reasonable ratio to provide a data foundation for model training and validation.

[0031] S42: Training the model. Based on the actual needs of crop water consumption prediction in the irrigation area, build an adapted Transformer model architecture and set the corresponding model parameters. Use the training set to iteratively train the Transformer model. During the training process, input the validation set data into the model in real time to monitor the model's fitting effect. Adjust the model parameters dynamically based on the feedback results of the validation set to avoid overfitting or underfitting. Stop training when the model's fitting effect reaches the preset requirements.

[0032] S43: Accuracy verification. Input the feature data from the test set into the trained Transformer model, and the model outputs the corresponding crop water consumption prediction value. Compare the prediction value with the actual crop water consumption data in the test set, calculate the error between the two using professional error calculation methods, and verify the prediction accuracy of the model. If the model accuracy reaches the preset application standard, the final crop water consumption prediction model that can be put into use is formed.

[0033] After the crop water consumption prediction model constructed in this embodiment is put into practical application, it only needs to input the future crop planting structure, planting time, basic soil data and meteorological forecast data of the irrigation area into the model. The model can quickly output the future water consumption prediction results of different plots and different crops in the irrigation area, realize high-precision and rapid prediction of crop water consumption in the irrigation area, provide data support for the implementation of water-saving irrigation and precision agricultural irrigation in the irrigation area, and can be widely used in the agricultural production practice of irrigation areas in various grain production areas and cash crop production areas.

[0034] The remaining technical features in the above embodiments can be flexibly selected by those skilled in the art to meet different specific practical needs according to actual circumstances. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims. In the above description, numerous specific details have been set forth to provide a thorough understanding of the present invention. However, it will be apparent to those skilled in the art that these specific details are not necessary to implement the present invention. In other instances, to avoid obscuring the present invention, well-known techniques, such as specific construction details, operating conditions, and other technical conditions, have not been specifically described.

[0035] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for predicting crop water consumption in an intelligent irrigation area, characterized in that, The steps are as follows: S1: Data collection, through a combination of data acquisition by agricultural management departments and remote sensing inversion, collects multi-source daily-scale data on crop water consumption in the irrigation area, including meteorological data, soil data, vegetation index and crop data; S2: Data processing, standardizing the collected multi-source data to achieve uniformity of data spatial resolution and precise alignment of time and spatial dimensions; S3: Crop water consumption estimation. Based on the processed standardized data, the daily evapotranspiration of crops is calculated using the principle of surface energy balance and the SEBAL model, which is used as the actual water consumption of crops. S4: Crop water consumption prediction model construction. A Transformer model is built as the core prediction model. Through dataset construction, model training and accuracy verification, a usable crop water consumption prediction model is obtained. Inputting planting structure, planting time, soil data and weather forecast data can output the prediction results of future crop water consumption.

2. The intelligent irrigation district crop water consumption prediction method according to claim 1, characterized in that: Step S1 specifically includes: S11: Meteorological data collection, collecting daily-scale observation data from meteorological stations within the irrigation area, including rainfall, surface temperature, air temperature, wind speed, and relative humidity; S12: Soil data collection, collecting soil data within the irrigation area, including soil salinity, soil type and soil texture; S13: Vegetation index extraction. The daily-scale vegetation indices of the irrigation area were extracted using Landsat 8 satellite sensors, including normalized vegetation index, enhanced vegetation index, green index, normalized water index, and soil-adjusted vegetation index. S14: Crop data collection, collecting crop data within the irrigation area, including crop planting time, growth and development time nodes, and planting structure data.

3. The intelligent irrigation district crop water consumption prediction method according to claim 1, characterized in that: Step S2 specifically includes: S21: Interpolation and resampling methods are used to unify all data to a spatial resolution of 30 meters; S22: Align all data in time, organize crop planting time into raster data with a spatial resolution of 30 meters, and spatially align it with planting structure data.

4. The intelligent irrigation district crop water consumption prediction method according to claim 1, characterized in that: Step S3 specifically includes: S31: Using remote sensing data and meteorological data, calculate net radiation, surface heat flux and sensible heat flux based on the principle of surface energy balance; S32: Calculate diurnal evapotranspiration based on the energy balance equation of the SEBAL model, using the following formula: in, Net radiation, For surface heat flux, For sensible heat flux, Evaporation rate It is the difference between the aerodynamic surface temperature and the reference altitude temperature. For surface temperature, and This is an empirical coefficient. It is air density. It is the specific heat of air. It is the aerodynamic drag that accounts for heat transfer between the aerodynamic surface and the reference altitude. For surface albedo, and For incident and outgoing longwave radiation, It is the surface thermal emissivity. It is the actual direct and diffuse solar radiation flux reaching the Earth's surface.

5. The intelligent irrigation district crop water consumption prediction method according to claim 1, characterized in that: Step S4, the construction of the crop water consumption prediction model, specifically includes the following sub-steps: S41: Construct a dataset, using the estimated crop water consumption as label data and meteorological data, soil data, and crop growth information as feature data, and divide them into training set, validation set, and test set. S42: Training the model. The Transformer model is trained using the training set, and the model parameters are adjusted in real time using the validation set to optimize the model's fitting effect. S43: Accuracy verification. Input the feature data from the test set into the trained Transformer model, output the predicted value of crop water consumption, calculate the error between the predicted value and the actual water consumption in the test set, verify the prediction accuracy of the model, and form the final crop water consumption prediction model.

Images