Crop irrigation decision-making method and system based on water demand characteristics in crop growth period

Abstract

The invention belongs to the field of crop growth monitoring, and particularly relates to a crop irrigation decision-making method and system based on crop growth period water demand characteristics, and the method comprises the steps: firstly obtaining a remote sensing satellite image of a to-be-decided crop soil region, and the segmented growth period and transition time period attributes of each crop; secondly, monitoring the soil water content of a crop soil area to be decided by using a temperature vegetation drought index according to a remote sensing satellite image; and finally, combining the soil water content with the related attributes of the crop growth period, and generating a variable irrigation prescription map by means of a preset variable precise irrigation decision model, so as to realize the irrigation decision of the crop soil area to be decided. By comprehensively utilizing the remote sensing data and the crop growth period information, the irrigation scheme can be reasonably planned according to the water demand characteristics of different crops in different growth stages, the water resource utilization efficiency is effectively improved, and the crop growth is guaranteed.

Description

Technical Field

[0001] The present invention belongs to the field of crop growth monitoring, and in particular relates to a crop irrigation decision-making method and system based on the water demand characteristics of crops during their growth period. Background Art

[0002] In modern agricultural production, irrigation is one of the key links to ensure high and stable crop yields. Traditional irrigation methods mostly rely on farmers' experience and intuition, and lack an accurate grasp of the actual water requirements of crops, which often leads to water waste or insufficient irrigation. With the intensification of global climate change and the increasing shortage of water resources, how to irrigate scientifically and rationally has become an important issue that needs to be addressed. Although existing technologies have introduced some automatic irrigation systems based on soil moisture, these systems usually fail to fully consider the specific water requirements of crops in different growth stages, making it difficult to achieve refined management. In addition, traditional irrigation decision-making models mostly use static parameter settings, which cannot adapt to complex changes in farmland environments and have limited prediction accuracy.

[0003] For example, a Chinese patent application with publication number CN116579872A discloses a precise irrigation decision-making method based on a crop growth model and weather forecast, including the steps of constructing a crop water requirement prediction model based on the crop growth model and constructing an irrigation decision-making model based on the crop growth model. The crop growth model is first corrected and verified based on historical data related to crop growth, and then different irrigation strategies are simulated for different crop growth models to determine the water requirement of water-saving and efficient crops in each growth period. Then, the real-time irrigation decision results of the crops in the study area are determined based on the day sequence number after crop sowing and the real-time weather forecast information.

[0004] For example, a Chinese patent with authorization announcement number CN110580657B discloses a method for predicting agricultural irrigation water demand, which includes selecting a base period, selecting crops with a sowing area greater than a set ratio as typical crops; calculating the irrigation water quota of typical crops during the growth period based on meteorological information; obtaining the ten-day rainfall at each meteorological station in the water-saving area, and calculating the effective rainfall of typical crops based on the ten-day rainfall; calculating the irrigation water demand per unit area of ​​typical crops; judging whether there is at least one typical crop whose irrigation water demand per unit area is greater than the average value of the actual irrigation water consumption per unit area in consecutive measured years, and if so, revising the irrigation water demand per unit area of ​​all typical crops; otherwise, using the irrigation water demand per unit area of ​​each typical crop and the planting area of ​​the next year to calculate the total water demand for the next year.

[0005] The above prior art has the following problems: When making irrigation decisions based on the water demand characteristics of crop growth stages, the prior art usually makes predictions based on the entire growth stage of the crop. However, crop growth is a dynamic process, and the water demand corresponding to each day is different. Therefore, it is impossible to accurately grasp the irrigation time intervals corresponding to different segmented growth stages of the crop, resulting in problems such as insufficient irrigation or over-irrigation during the corresponding segmented growth stages. Therefore, the present invention provides a crop irrigation decision-making method and system based on the water demand characteristics of crop growth stages. Summary of the Invention

[0006] In view of the deficiencies of the prior art, the present invention proposes a crop irrigation decision-making method and system based on the water demand characteristics of crop growth stages, including: constructing a standard water demand space based on the attributes of segmented crop growth stages and standard growth water demands, and combining the corresponding weather, evaporation, deep percolation, and crop absorption rates to construct a water demand dynamic balance prediction equation, and then constructing a two-way water demand prediction model to obtain the crop water demand, irrigation time interval, and water demand deviation factor, and constructing an irrigation trigger interval, irrigation time interval, and irrigation intensification decision model based on this to generate an irrigation strategy, and monitoring and evaluating the crop growth and water storage status, and feeding back the evaluation results to the prediction model and decision model to adjust the predicted water storage and irrigation strategy in real time; the present invention can accurately calculate the water demand and generate an irrigation decision based on the water demand characteristics of the crop growth stage, effectively improving the irrigation efficiency and water resource utilization rate, and ensuring the good growth state of the crop.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A crop irrigation decision-making method based on the water demand characteristics of crop growth stages, including:

[0009] Obtaining remote sensing satellite images of the soil area of the crop to be decision-making and the attributes of each segmented crop growth stage and transition time period;

[0010] Monitoring the soil water content of the soil area of the crop to be decision-making according to the remote sensing satellite image and the temperature vegetation drought index;

[0011] Generating a variable irrigation prescription map according to the soil water content, the attributes of each segmented crop growth stage and transition time period, and combining a preset variable precision irrigation decision-making model to make a decision on the crop irrigation in the soil area of the crop to be decision-making.

[0012] Specifically, monitoring the soil water content of the soil area of the crop to be decision-making includes:

[0013] Inverting the soil water content, temperature vegetation drought index, surface temperature, and normalized difference vegetation index under drought stress according to the remote sensing satellite image;

[0014] Predict the soil water content in the soil area of the crop to be determined based on the measured soil water content, the soil water content under drought stress, the temperature vegetation drought index, the land surface temperature, and the normalized difference vegetation index.

[0015] Specifically, the variable precision irrigation decision-making model is obtained by integrating the water demand two-way prediction model and the irrigation intensification decision-making model; the steps of generating a variable irrigation prescription map in combination with the preset variable precision irrigation decision-making model include:

[0016] According to the segmented growth period and transitional time period attributes of each crop and the soil water content, through the water demand two-way prediction model constructed and pre-trained, obtain the water demand and irrigation time interval of the corresponding type of crop;

[0017] Obtain the water demand deviation factor according to the water tolerance degree of each segmented growth period of the corresponding type of crop;

[0018] According to the water demand, irrigation time interval and water demand deviation factor of the corresponding type of crop, through the configured irrigation intensification decision-making model, generate a variable irrigation prescription map for the soil area of the crop to be determined.

[0019] Specifically, the segmented growth period and transitional time period attributes of each crop include: the corresponding meteorological data, evaporation state, deep soil infiltration state, crop absorption rate during the growth period, the standard water demand space in the kth growth period, and the meteorological data, evaporation state, soil moisture, and crop absorption rate data in the pre-transition section and post-transition section within the kth growth period;

[0020] The steps for constructing and training the water demand two-way prediction model include:

[0021] S301. Set the pre-transition section and post-transition section for each growth period stage of the crop to be determined, and construct a transition change function according to the historical water demand corresponding to the pre-transition section and post-transition section;

[0022] The pre-transition section is a time period that advances the first preset number of days before the corresponding growth period stage, and the post-transition section is a time period that advances the first preset number of days after the corresponding growth period stage;

[0023] S302. Based on the BP neural network and the particle swarm optimization algorithm, construct K water demand prediction sub-models integrated in the water demand two-way prediction model, embed the transition change function into the water demand prediction sub-model corresponding to the adjacent growth period, and embed the crop water demand dynamic balance prediction equation corresponding to the growth period into the water demand prediction sub-model corresponding to the growth period;

[0024] S303. Input the corresponding meteorological data, evaporation state, soil moisture, crop absorption rate, standard water requirement space in the k-th growth period, and meteorological data, evaporation state, soil moisture, crop absorption rate data in the pre-transition section and post-transition section of the k-th growth period, as well as the soil water content of the soil area of the crop to be determined into the water requirement prediction sub-model corresponding to the k-th growth period for training, and obtain the transition section prediction training error and non-transition section training error corresponding to the K water requirement prediction sub-models;

[0025] S304. Set a training error threshold, construct a comprehensive training loss function based on the obtained K transition section prediction training errors and K - 1 non-transition section training errors, and obtain the trained water requirement prediction sub-model when the comprehensive training loss function is less than the training error threshold.

[0026] Specifically, the steps for obtaining the transition section prediction training error and non-transition section training error include:

[0027] S3031. When the water requirement prediction time point at the current moment in the k-th growth period is in the non-transition section of the current growth period, predict and obtain the water requirement at the corresponding time point through the water requirement prediction sub-model corresponding to the k-th growth period, and use the predicted water requirement and the historical true water requirement to obtain the k-th non-transition section training error;

[0028] S3032. When the water requirement prediction time point at the current moment in the k-th growth period is in the pre-transition section of the current growth period, perform a pre-prediction through the water requirement prediction sub-model corresponding to the k-th growth period to obtain the predicted water requirement at time t in the pre-transition section of the k-th growth period, and at the same time, predict and obtain the predicted water requirement at the same time point through the water requirement prediction sub-model corresponding to the (k - 1)-th growth period.

[0029] Specifically, the steps for obtaining the transition section prediction training error and non-transition section training error further include:

[0030] S3033. Use the average value of the predicted water requirement obtained by the water requirement prediction sub-model corresponding to the k-th growth period and the predicted water requirement obtained by the water requirement prediction sub-model corresponding to the (k - 1)-th growth period at the same time point as the predicted water requirement corresponding to time t in the pre-transition section, and use the predicted water requirement corresponding to time t and the water requirement at the corresponding time point to obtain the transition section prediction training error corresponding to the time point in the pre-transition section;

[0031] S3034. When the water requirement prediction time point at the current moment in the k-th growth period is in the post-transition section of the current growth period, repeat the process of S3032 to obtain the predicted water requirement at time t in the post-transition section for the water requirement prediction sub-model corresponding to the k-th growth period and the predicted water requirement at the same time point for the water requirement prediction sub-model corresponding to the (k + 1)-th growth period. Repeat S3033 to obtain the prediction training error of the transition section at the corresponding time point in the post-transition section.

[0032] Specifically, the steps for obtaining the water requirement deviation factor include:

[0033] S311. Configure the water tolerance experiment for different growth stages of the crop, and set the water requirement obtained by predicting the two-way water requirement model for the corresponding growth stage of the crop as the ideal water requirement.

[0034] S312. Starting from the ideal water requirement, increase or decrease the water requirement of the crop in the corresponding growth period, monitor the growth state of the crop corresponding to the water requirement at each moment, and automatically evaluate the growth state score of the crop under the corresponding water requirement using the configured evaluation algorithm.

[0035] S313. Using the water requirement and the growth state score under the corresponding water requirement, obtain the water requirement-growth score change curve corresponding to different growth stages of the corresponding crop throughout the growth period.

[0036] S314. Starting from the growth state score of the crop corresponding to the ideal water requirement in each growth stage, when increasing or decreasing the water requirement and the growth state scores corresponding to the water requirement are continuously less than the growth state score of the crop corresponding to the ideal water requirement for m times, obtain the minimum water requirement and the maximum water requirement corresponding to the start of the decline in the growth state score of the crop in the corresponding growth stage.

[0037] S315. According to the minimum water requirement and the ideal water requirement, obtain the lower limit deviation factor of the water requirement for the corresponding growth stage, and at the same time, according to the maximum water requirement and the ideal water requirement, obtain the upper limit deviation factor of the water requirement for the corresponding growth stage.

[0038] Specifically, the steps for constructing the irrigation intensification decision model include:

[0039] S401. Based on the ideal water requirement, the lower limit deviation factor of the water requirement, and the upper limit deviation factor of the water requirement, construct the water requirement interval for each growth stage of the corresponding crop.

[0040] S402. Use the water requirement interval for each growth stage as the variable irrigation prescription map, and set the trigger information for the irrigation strategy execution action according to the variable irrigation prescription map.

[0041] Specifically, the steps for setting the trigger information for the irrigation strategy execution action include:

[0042] S4021. When the water holding capacity of the crop growth area corresponding to the current moment is lower than the lower limit of the water demand range in the variable irrigation prescription map, direct irrigation is carried out until the ideal water demand for the corresponding growth stage is reached;

[0043] S4022. When the water holding capacity of the crop growth area corresponding to the current moment is lower than the ideal water demand for the corresponding growth stage, the parameters of the irrigation intensification decision model are adjusted in advance according to the ideal water demand for the corresponding growth stage. When the water holding capacity of the corresponding crop growth area reaches the lower limit of the water demand range in the variable irrigation prescription map, the irrigation device is controlled to irrigate through the pre-adjusted parameters of the irrigation intensification decision model until the ideal water demand for the corresponding growth stage is reached;

[0044] S4023. When the water holding capacity of the crop growth area corresponding to the current moment is higher than the ideal water demand for the corresponding growth stage, no irrigation is carried out. When it is lower than the ideal water demand for the corresponding growth stage, the process of S4022 is repeated.

[0045] A crop irrigation decision-making system based on the water demand characteristics of crop growth stages includes: a prediction module and an irrigation decision-making module.

[0046] The prediction module is used to monitor the soil water content in the soil area of the crop to be decided according to the remote sensing satellite image and the temperature vegetation drought index;

[0047] The irrigation decision-making module is used to generate a variable irrigation prescription map according to the soil water content, the segmented growth stages of each crop and the attributes of the transition time period, and combine with a preset variable precision irrigation decision-making model to make decisions on crop irrigation in the soil area of the crop to be decided.

[0048] Compared with the prior art, the beneficial effects of the present invention are:

[0049] Aiming at the deficiencies of the prior art, the present invention accurately predicts the soil water content by obtaining remote sensing satellite images and attributes such as the segmented growth stages of crops; uses a uniquely constructed and trained two-way water demand prediction model, combines the weather, evaporation, soil moisture, absorption rate and other states of each growth stage of the crop, and can accurately determine the water demand and irrigation time interval of different segmented growth stages, avoiding the problems of insufficient or excessive irrigation; at the same time, according to the water tolerance degree, a water demand deviation factor is obtained, and a variable irrigation prescription map is generated through the irrigation intensification decision model, realizing dynamic and accurate irrigation decision-making; not only improving the water resource utilization efficiency and reducing waste, but also providing suitable moisture for the crop at different growth stages, effectively guaranteeing the healthy growth of the crop. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a flow chart of the crop irrigation decision-making method based on the water demand characteristics of crop growth stages in Embodiment 1 of the present invention;

[0051] Figure 2 This is the display diagram of the non-transition section and the transition section in Embodiment 1 of the present invention;

[0052] Figure 3 This is the predicted water requirement recommended depth prescription diagram corresponding to the rice planting plot in Embodiment 1 of the present invention;

[0053] Figure 4 This is the module diagram of the crop irrigation decision-making system based on the water requirement characteristics of crop growth periods in Embodiment 2 of the present invention. Detailed implementation manners

[0054] Embodiment 1

[0055] Please refer to Figure 1 , an embodiment provided by the present invention: a crop irrigation decision-making method based on the water requirement characteristics of crop growth periods, the steps include:

[0056] S1. Obtain the remote sensing satellite image of the soil area of the crop to be decision-making and the attributes of each crop's segmented growth period and transition time period;

[0057] S2. Monitor the soil water content of the soil area of the crop to be decision-making according to the remote sensing satellite image and the temperature vegetation drought index;

[0058] Further, predicting the soil water content of the soil area of the crop to be decision-making in this embodiment includes:

[0059] Inverting the soil water content, temperature vegetation drought index, surface temperature, and normalized difference vegetation index under drought stress according to the remote sensing satellite image;

[0060] Further, the specific steps of inverting the soil water content under drought stress in this embodiment include:

[0061] Inverting through a simulation statistical model according to the remote sensing satellite image, and using the measured soil water and the corresponding VTCI value to fit a linear equation of VTCI and soil water, and obtaining the soil water content of the soil area of the crop to be decision-making according to the linear equation of VTCI and soil water;

[0062] Further, the steps of obtaining the temperature vegetation drought index in this embodiment include:

[0063] Extracting the near-infrared band and the red light band according to the remote sensing image monitored in the area to be irrigated;

[0064] Obtaining the NDVI value of each pixel in the area to be irrigated according to the reflectance values of the near-infrared band and the red light band;

[0065] where NDVI represents the normalized vegetation index;

[0066] According to the remotely sensed image of the area to be irrigated, two thermal infrared bands are extracted, and through the split-window algorithm, the true land surface temperature value (LST, Land Surface Temperature) is obtained;

[0067] According to the NDVI value of each pixel and the corresponding true land surface temperature value, with NDVI as the abscissa and LST as the ordinate, a two-dimensional LST-NDVI feature space is constructed;

[0068] Analyze the scatter points in the LST-NDVI feature space to determine the dry edge and the wet edge;

[0069] According to the LST-NDVI feature space and the dry edge and the wet edge, the drought index of each pixel is obtained, specifically:

[0070] ,

[0071] Among them, is the land surface temperature corresponding to the i-th pixel with the normalized vegetation index of ; is the maximum land surface temperature of the i-th normalized vegetation index in the region; is the minimum land surface temperature of the i-th normalized vegetation index in the region.

[0072]

[0073] Among them, the equation coefficients and represent the corresponding equation coefficients, which can be obtained by fitting the equation of the dry-wet edge scatter plot of the "LST-NDVI feature space".

[0074] According to the measured soil water content, soil water content under drought stress, temperature vegetation drought index, land surface temperature, and normalized difference vegetation index, the soil water content of the soil area of the crop to be determined is inverted.

[0075] S3. According to the soil water content, the segmented growth period and transition time period attributes of each crop, and in combination with a preset variable precision irrigation decision model, a variable irrigation prescription map is generated to make a decision on the crop irrigation in the soil area of the crop to be determined.

[0076] Furthermore, the variable precision irrigation decision model in this embodiment is obtained by integrating a water demand two-way prediction model and an irrigation strengthening decision model; the step of generating a variable irrigation prescription map in combination with the preset variable precision irrigation decision model includes:

[0077] According to the segmented growth periods of each type of crop, the attributes of the transition time periods, and the soil water content, through the constructed and pre-trained two-way water demand prediction model, obtain the water demand and irrigation time intervals of the corresponding type of crop;

[0078] Obtain the water demand deviation factor according to the water tolerance degree of each segmented growth period of the corresponding type of crop;

[0079] According to the water demand of the corresponding type of crop, the irrigation time interval, and the water demand deviation factor, generate a variable irrigation prescription map for the soil area of the crop to be decision-making through the configured irrigation reinforcement decision model.

[0080] Furthermore, in this embodiment, the attributes of the segmented growth periods of each type of crop and the transition time periods include: the corresponding meteorological data, evaporation state, soil moisture condition, crop absorption rate during the growth period, the standard water demand space in the kth growth period, and the meteorological data, evaporation state, soil moisture condition, and crop absorption rate data in the pre-transition section and the post-transition section within the kth growth period; the soil moisture condition includes the soil deep infiltration state and the water infiltration rate corresponding to the growth stage of the crop growth area;

[0081] Furthermore, the construction and training steps of the two-way water demand prediction model in this embodiment include:

[0082] S301. Set the pre-transition section and the post-transition section corresponding to each growth period stage of the crop to be decision-making, and construct a transition change function according to the historical water demands corresponding to the pre-transition section and the post-transition section;

[0083] Furthermore, in this embodiment, the pre-transition section is the time period that advances the first preset number of days before the corresponding growth period stage, and the post-transition section is the time period that advances the first preset number of days after the corresponding growth period stage; furthermore, the first preset number of days in this embodiment is set by those skilled in the art according to the corresponding crop type and crop growth attributes.

[0084] S302. Based on the BP neural network and the particle swarm optimization algorithm, construct K water demand prediction sub-models integrated in the two-way water demand prediction model, embed the transition change function into the water demand prediction sub-model corresponding to the adjacent growth periods, and embed the crop water demand dynamic balance prediction equation corresponding to the growth period into the water demand prediction sub-model corresponding to the growth period;

[0085] Furthermore, the construction of the crop water demand dynamic balance prediction equation in this embodiment includes:

[0086] ,

[0087] Among them, represents the water demand change amount corresponding to the ith type of crop at the t moment in the kth growth stage, Denote the cumulative average rainfall corresponding to the crop's corresponding growth area at the t-th moment of the k-th growth stage. Denote the irrigation water requirement corresponding to the i-th crop at the t-th moment of the k-th growth stage in the corresponding growth area. Denote the combined evapotranspiration corresponding to the i-th crop at the t-th moment of the k-th growth stage in the corresponding growth area; Denote the surface runoff corresponding to the crop at the t-th moment of the k-th growth stage in the corresponding growth area; Denote the effective infiltration factor corresponding to the i-th crop at the t-th moment of the k-th growth stage in the corresponding growth area; Denote the reference evapotranspiration corresponding to the crop at the t-th moment of the k-th growth stage in the corresponding growth area; Denote the crop coefficient of the i-th crop corrected by the radiation intensity, crop leaf surface area, and radiation angle at the corresponding time point at the k-th growth stage in the corresponding growth area; Denote the water absorption rate corresponding to the i-th crop at the k-th growth stage in the corresponding growth area; Denote the water evaporation corresponding to the crop growth area at the k-th growth stage; Denote the average effective root depth of the i-th crop at the k-th growth stage in the growth area, that is, the root depth of the crop for water absorption; Denote the infiltration amount corresponding to the i-th crop at the t-th moment of the k-th growth stage in the corresponding growth area; Denote the average infiltration amount of the i-th crop at the k-th growth stage in the corresponding growth area; Denote the water infiltration rate corresponding to the crop growth area at the k-th growth stage; Denote the standard growth water requirement corresponding to the i-th crop at the k-th growth stage in the crop growth area; Further, Denote the average water infiltration depth of the i-th crop at the k-th growth stage in the crop growth area;

[0088] In this embodiment, multiple factors are of great significance for crop water requirement calculation and irrigation decision-making. After deducting runoff loss from the effective infiltration amount, it affects the calculation of the change in water requirement. When the soil texture is good, its value is large, which can reduce irrigation demand and maintain soil balance. The effective infiltration factor is multiplied by the infiltration amount and corrected according to the crop and growth stage. During the late growth stage of crops with well-developed roots, this factor is large, which can more accurately reflect the soil water utilization of different crops at each stage. The crop coefficient varies according to the crop type and growth stage, and is large during the vigorous growth period, such as the booting stage of wheat, etc., which affects the combined evapotranspiration amount and the change in water requirement, and helps to accurately adapt to the evapotranspiration demand. The water absorption rate affects the change in water requirement. It is large during the vigorous growth period and requires more irrigation, and is small during the slow growth or stress period, so the dynamic water requirement can be grasped accordingly to achieve precise irrigation. The average effective root depth affects the calculation of the effective infiltration amount. In the initial stage, it is shallow and only shallow water can be used. During growth, it deepens and deep water can be used to optimize the decision-making of irrigation depth and amount. The water infiltration rate and the average effective root depth jointly affect the infiltration amount and the effective infiltration amount. A fast infiltration rate is conducive to water replenishment and storage, which can reduce surface runoff after rain or irrigation, improve the irrigation effect and promote crop growth. The combined action of these factors helps to accurately predict crop water requirements, optimize irrigation strategies, and improve water resource utilization efficiency.

[0089] S303. Input the corresponding meteorological data, evaporation state, soil moisture, crop absorption rate, the standard water requirement space in the k-th growth period, and the meteorological data, evaporation state, soil moisture, crop absorption rate data in the pre-transition section and post-transition section in the k-th growth period, as well as the soil water content in the soil area of the crop to be decision-making into the water requirement prediction sub-model corresponding to the k-th growth period for training, and obtain the transition section prediction training error and non-transition section training error corresponding to the K water requirement prediction sub-models;

[0090] Furthermore, in this embodiment, the construction steps of the standard water requirement space for crop growth periods include:

[0091] S101. According to the biological characteristics and growth laws of crops, divide the growth periods of each crop into segments; for example, for wheat, it can be divided into sowing-emergence stage, seedling stage, jointing stage, booting stage, heading stage, filling stage, and maturity stage, etc.;

[0092] S102. Refer to agricultural science literature, crop cultivation manuals, and long-term field observation records to determine the start and end times and landmark features of each growth period, and these features can include plant morphological changes, physiological index changes, etc.;

[0093] S103. Obtain data through field experiments, that is, in different experimental fields, while controlling other environmental factors to be relatively stable, measure the actual water requirements of each crop during each sub - growth period; further, in this embodiment, the water balance method is adopted, that is, measure irrigation water volume, precipitation, and soil moisture change amount, etc. to calculate the actual water consumption of the crop during this growth period, and use this as a reference for the standard growth water requirement;

[0094] S104. Collect historical agricultural data, including agricultural irrigation records, meteorological station data, and crop yield data for many years. Use statistical analysis methods to extract the average water requirement information of each crop during each growth period from these data; at the same time, refer to the crop water requirement standard data released by agricultural scientific research institutions;

[0095] S105. Use a three - dimensional space model (growth period, water requirement, environmental factors) or a multi - dimensional matrix model to represent the standard water requirement space of the crop's sub - growth period; for example, take the growth period as one dimension and the standard growth water requirement as another dimension, establish a two - dimensional matrix, where the rows represent different growth periods and the columns represent different water requirement ranges or specific values; and introduce some auxiliary variables, such as soil type, variety differences, etc.; because different soil water - retaining capacities and crop varieties have different water requirements and utilization efficiencies, these variables can be used as adjustment parameters of the model to make the construction of the standard water requirement space more accurate;

[0096] S106. Fill the collected crop sub - growth period attribute data and standard growth water requirement data into the established model framework. For each growth period, fill the corresponding standard growth water requirement value or range into the corresponding position; for example, in the two - dimensional matrix, fill the standard growth water requirement range of the wheat seedling stage into the corresponding row;

[0097] S107. Calibrate the filled model by comparing data from different sources (field experiment data, historical data, and standard data). If large errors are found between the data, analyze the error reasons and perform error optimization and calibration.

[0098] Further, the steps for obtaining the transition segment prediction training error and non - transition segment training error in this embodiment include:

[0099] S3031. When the water requirement prediction time point at the current moment of the k - th growth period is in the non - transition segment of the current growth period, predict and obtain the water requirement at the corresponding time point through the water requirement prediction sub - model corresponding to the k - th growth period, and use the predicted water requirement and the historical true water requirement to obtain the k - th non - transition segment training error;

[0100] Further, please refer to Figure 2, where the part labeled 1 is the pre-transition section corresponding to the k-th growth stage, and the part labeled 2 is the post-transition section corresponding to the (k - 1)-th growth stage. In this embodiment, the transition sections corresponding to labels 1 and 2 constitute the complete adjacent growth stage transition section between the (k - 1)-th growth stage and the k-th growth stage. The part labeled 3 is the non-transition section part corresponding to the k-th growth stage, and the other black parts in the figure are the same as the above description;

[0101] S3032. When the water demand prediction time point at the current moment of the k-th growth stage is in the pre-transition section of the current growth stage, perform pre-prediction through the water demand prediction sub-model corresponding to the k-th growth stage to obtain the predicted water demand at time t corresponding to the pre-transition section of the k-th growth stage. At the same time, predict the predicted water demand at the same time through the water demand prediction sub-model corresponding to the (k - 1)-th growth stage;

[0102] S3033. Use the average value of the predicted water demand obtained by the water demand prediction sub-model corresponding to the k-th growth stage and the predicted water demand obtained at the same time by the water demand prediction sub-model corresponding to the (k - 1)-th growth stage as the predicted water demand corresponding to time t in the pre-transition section, and use the predicted water demand corresponding to time t and the water demand at the corresponding time point to obtain the transition section prediction training error at the corresponding time point of the pre-transition section;

[0103] Further, in this embodiment, the specific process of using the average value of the predicted water demand obtained by the water demand prediction sub-model corresponding to the k-th growth stage and the predicted water demand obtained at the same time by the water demand prediction sub-model corresponding to the (k - 1)-th growth stage as the predicted water demand corresponding to time t in the pre-transition section is as follows: Assume that at time t in the pre-transition section 1, use the water demand prediction sub-model corresponding to the k-th growth stage and the water demand prediction sub-model corresponding to the (k - 1)-th growth stage to predict the water demand respectively, and then take the average value of the predicted water demand obtained corresponding to the k-th growth stage and the predicted water demand obtained corresponding to the (k - 1)-th growth stage as the predicted water demand at time t.

[0104] S3034. When the water demand prediction time point at the current moment of the k-th growth stage is in the post-transition section of the current growth stage, repeat the process of S3032 to obtain the predicted water demand at time t corresponding to the post-transition section by the water demand prediction sub-model corresponding to the k-th growth stage and the predicted water demand at the same time corresponding to the water demand prediction sub-model corresponding to the (k + 1)-th growth stage, and repeat S3033 to obtain the transition section prediction training error at the corresponding time point of the post-transition section;

[0105] Further, in this embodiment, when the water demand prediction time point at the current moment is in the transition period of the current growth stage, the water demand at each time point of the corresponding transition period is obtained by the average of the backward sliding prediction of the previous growth stage and the forward sliding prediction of the next growth stage, which can accurately integrate the crop attribute characteristics of two adjacent periods and reduce the influence of the change of the crop growth stage corresponding to a single growth stage on the water demand prediction.

[0106] S304. Set a training error threshold, construct a comprehensive training loss function according to the obtained training errors of K transition segments and the training errors of K - 1 non-transition segments, and obtain the trained water demand prediction sub-model when the comprehensive training loss function is less than the training error threshold.

[0107] This process first sets the pre-transition and post-transition segments and constructs a transition change function, which can better reflect the gradual change process of the water demand of crops during the growth stage transition and conforms to the crop growth law. The adjacent growth stage water demand prediction sub-models are used for joint prediction of the transition segments, integrating the crop attribute characteristics of different growth stages, reducing the influence of the growth stage transition on the water demand prediction, and improving the prediction accuracy; the model is constructed based on the BP neural network and the particle swarm algorithm, which can effectively process the complex relationships of multiple variables and optimize the model performance; by comparing the predicted water demand with the historical real water demand to calculate the training error, the model parameters can be adjusted in time to make the model continuously approach the real situation. The finally obtained trained model can accurately predict the crop water demand and the irrigation time interval, providing strong support for scientific irrigation.

[0108] Further, the steps for obtaining the water demand deviation factor in this embodiment include:

[0109] S311. Configure the water tolerance experiment for different growth stages of the crop, and set the water demand of the crop corresponding to the growth stage predicted by the two-way water demand prediction model as the ideal water demand;

[0110] S312. Starting from the ideal water demand, increase or decrease the water demand of the crop in the corresponding growth stage, monitor the growth state of the crop corresponding to the water demand at each moment, and automatically evaluate the growth state score of the crop under the corresponding water demand by using the configured evaluation algorithm;

[0111] S313. Use the water demand and the growth state score under the corresponding water demand to obtain the water demand-growth score change curve corresponding to different growth stages of the corresponding crop throughout the growth period;

[0112] S314. Starting from the growth state score of the crop corresponding to the ideal water demand in each growth stage, when the growth state score corresponding to the increased or decreased water demand is continuously less than the growth state score of the crop corresponding to the ideal water demand for m times, obtain the minimum water demand and the maximum water demand corresponding to the start of the decline of the growth state score of the crop in the corresponding growth stage;

[0113] S315. Obtain the lower limit deviation factor of water requirement for the corresponding growth stage according to the minimum water requirement and the ideal water requirement, and at the same time, obtain the upper limit deviation factor of water requirement for the corresponding growth stage according to the maximum water requirement and the ideal water requirement.

[0114] Furthermore, the calculation process of the water requirement deviation factor in this embodiment is quantified based on the relative difference between the ideal water requirement of the crop and the actual moisture tolerance critical value. Specifically, when determining the lower limit deviation factor of a certain growth stage, first, based on the ideal water requirement of this stage, subtract the minimum water requirement measured in the experiment (i.e., the minimum water requirement corresponding to the first continuous decline in the crop growth state score), and then divide the difference between the two by the ideal water requirement. The resulting ratio is the maximum relative threshold for the allowable reduction of water requirement in this stage. Similarly, the upper limit deviation factor is calculated by dividing the difference between the critical maximum water requirement (the maximum water requirement before the continuous decline in the growth state score) and the ideal water requirement by the ideal water requirement, which reflects the maximum allowable increase ratio of water requirement that the crop can withstand. This method converts the absolute difference between the critical water requirement and the ideal value into a normalized ratio based on the ideal value, eliminates the interference of the absolute value difference of water requirement in different growth stages, constructs a dimensionless dynamic tolerance interval parameter, and thus accurately depicts the response boundary of the crop to water fluctuations, providing a unified and comparable deviation control benchmark for the dynamic adaptation of the irrigation strategy across growth stages.

[0115] In this process, determine the irrigation thresholds for different growth stages according to the calculation results of the two-way prediction model of crop water requirement and the tolerance degree of crop growth to water stress; the irrigation thresholds include the lower limit threshold and the upper limit threshold. When the lower limit threshold is triggered, it means that the soil moisture has dropped to a level that may affect the normal growth of the crop, and irrigation is needed to supplement water; when the upper limit threshold is reached, it indicates that the soil moisture is close to saturation or may have an adverse impact on the crop due to excessive water (such as root hypoxia, etc.), and irrigation should be stopped. For example, during the jointing stage of wheat, according to previous research and model prediction, the lower limit threshold of soil volumetric water content is determined to be 60% of the field capacity, and the upper limit threshold is 85% of the field capacity.

[0116] This process determines the ideal water requirement by configuring the moisture tolerance experiment and combining the prediction model, providing a benchmark for subsequent analysis. Based on the ideal water requirement, adjust and monitor the crop growth state and score, and accurately draw the change curve of water requirement - growth score, visually presenting the relationship between water requirement and growth state. The determined lower limit and upper limit deviation factors of water requirement clarify the tolerable water requirement range of the crop in each growth stage. This helps to achieve precise irrigation, avoid affecting crop growth due to insufficient or excessive irrigation, ensure the growth and development of the crop under suitable moisture conditions, improve the yield and quality of the crop, and at the same time achieve the efficient utilization of water resources.

[0117] Further, the irrigation strategy for each crop corresponding to the segmented growth period in this embodiment includes detailed information such as the start and end times of irrigation, the irrigation duration, the irrigation water volume, and the irrigation area;

[0118] Further, the steps for constructing the irrigation intensification decision model in this embodiment include:

[0119] S401. Based on the ideal water requirement, the lower limit deviation factor of the water requirement, and the upper limit deviation factor of the water requirement, construct the water requirement interval for each growth period stage of the corresponding crop;

[0120] S402. Use the water requirement interval for each growth period stage as the variable irrigation prescription map for the crop, and set the trigger information for the irrigation strategy execution action according to the variable irrigation prescription map. Specifically:

[0121] S4021. When the water holding capacity of the corresponding crop growth area at the current moment is lower than the lower limit of the variable irrigation prescription map, directly perform irrigation until the ideal water requirement for the corresponding growth period stage is reached;

[0122] Further, when the corresponding irrigated crop is rice, the comparison here is: when the water holding capacity of the corresponding crop growth area at the current moment is lower than the water depth of the plot corresponding to the lower limit of the variable irrigation prescription map, directly perform irrigation until the ideal water requirement depth for the corresponding growth period stage of rice is reached;

[0123] Further, in this embodiment, the corresponding variable irrigation prescription map includes the cumulative average rainfall, that is, meteorological data. When the weather changes, resulting in changes in the water requirement interval of the corresponding plot or the depth interval corresponding to the paddy field, the corresponding variable irrigation prescription map or the change interval of the plot water depth changes simultaneously, so as to achieve the purpose of adjusting the water requirement of the corresponding plot according to the weather conditions.

[0124] Further, the water holding capacity of the crop growth area in this embodiment is the water holding amount corresponding to the crop monitored at the current moment;

[0125] S4022. When the water holding capacity of the corresponding crop growth area at the current moment is lower than the ideal water requirement for the corresponding growth period stage, pre-adjust the parameters of the irrigation intensification decision model according to the ideal water requirement for the corresponding growth period stage. When the water holding capacity of the corresponding crop growth area reaches the lower limit of the variable irrigation prescription map, control the irrigation device to perform irrigation through the pre-adjusted parameters of the irrigation intensification decision model until the ideal water requirement for the corresponding growth period stage is reached;

[0126] S4023. When the water holding capacity of the corresponding crop growth area at the current moment is higher than the ideal water requirement for the corresponding growth period stage, do not perform irrigation. When it is lower than the ideal water requirement for the corresponding growth period stage, repeat the process of S4022;

[0127] In this embodiment, the water requirement intervals corresponding to different time points during the growth periods of different crops are obtained through the above process. If the crop to be irrigated is rice, then according to the obtained water requirements, based on the corresponding plot area, the water depth of the plot at the current moment, and the soil moisture condition of the plot, it is converted into a prescription map of the water requirement depth interval for the corresponding plot. For example, Figure 3 in Figure 3 , A-1 to A-5, B-1 to B-5, C-1 to C-5, and D-1 to D-5 respectively correspond to different rice plots in the current growth stage and the predicted rice water requirement depth intervals for the corresponding plots. However, in the actual irrigation and display process, the average value of the prescription map of the water requirement depth interval for the corresponding plot is used as the actual water requirement depth for the corresponding rice plot for real-time recommended irrigation and display. For example, Figure 3 in Figure 3 , the average recommended irrigation depths corresponding to the prescription maps of the water requirement depth intervals for the plots with different colors in the legend include 8mm, 10mm, 12mm, 14mm, etc.

[0128] If it is other dryland crops, such as wheat, corn, etc., then the corresponding water requirement intervals are directly used as the variable irrigation prescription map for irrigation control;

[0129] The variable irrigation prescription maps corresponding to each stage of the growth period of each crop are input into the configured irrigation device for irrigation, and the growth status and water storage status of the crops in the corresponding irrigation area after being irrigated by the irrigation strategy are monitored and evaluated in real time, and the evaluation results are fed back to the two-way water requirement prediction model and the irrigation reinforcement decision model to adjust the predicted water storage and irrigation strategy in real time. Further, this process uses the configured comprehensive fuzzy evaluation algorithm to evaluate the growth status of the crops in real time.

[0130] In this embodiment, the irrigation strategy is sent to the automatic irrigation control subsystem, which can be an intelligent irrigation system based on PLC (Programmable Logic Controller) or other types of automatic irrigation equipment; the automatic irrigation control subsystem accurately controls operations such as the opening and closing of irrigation valves, the start and stop of water pumps, and the operation of sprinklers according to the received instructions, realizing automatic precise irrigation.

[0131] After each irrigation is completed, continue to use the soil moisture sensor to monitor the change of soil moisture, and combine the monitoring data of crop growth indicators (such as plant height, leaf area, biomass, etc.) to evaluate the effect of this irrigation. For example, observe whether the soil moisture has recovered within a reasonable range and remained stable after irrigation, and whether the crop growth has been promoted by irrigation and there are no abnormal growth phenomena caused by too much or too little water.

[0132] Based on the evaluation results of irrigation effects, feedback adjustments are made to the two-way prediction model of crop water requirements, irrigation thresholds, and parameters in the irrigation plan. If it is found that the soil moisture recovery rate is too slow or too fast after a certain irrigation, it may be necessary to appropriately adjust the calculation parameters of irrigation water volume or irrigation time; if abnormal crop growth occurs, such as yellowing leaves, stunted growth, etc., it may be necessary to re-examine whether the irrigation threshold is reasonable and optimize it in combination with the two-way prediction model of crop water requirements and actual conditions. Through continuous evaluation and feedback adjustment, the irrigation decision-making method can better adapt to the crop irrigation needs under different years, different seasons, and different field conditions.

[0133] Furthermore, in this embodiment, a variable irrigation prescription map and an irrigation reinforcement decision model are constructed. By setting action trigger information based on the water requirement interval, precise irrigation can be achieved. Whether to irrigate and the irrigation method are determined by comparing the water volume in the crop growth area with the trigger interval, avoiding over-irrigation or under-irrigation and improving the water resource utilization efficiency. The irrigation strategy is applied to the actual situation and monitored and evaluated in real time, and the feedback results are used to adjust the prediction and decision models, continuously optimizing the models to make them more in line with the actual situation. Precise irrigation helps to create a suitable growth environment for crops, promote crop growth and development, and improve crop yield and quality. At the same time, this dynamic adjustment mechanism can adapt to environmental changes and crop growth dynamics, enhancing the flexibility and adaptability of the irrigation system.

[0134] Embodiment 2

[0135] Please refer to Figure 4 , another embodiment provided by the present invention: a crop irrigation decision-making system based on the water requirement characteristics of the crop growth period, including: a prediction module and an irrigation decision module.

[0136] The prediction module is used to monitor the soil water content in the soil area of the crop to be decided according to the remote sensing satellite image and the temperature vegetation drought index;

[0137] The irrigation decision module is used to generate a variable irrigation prescription map according to the soil water content, the segmented growth period and the transition time period attributes of each crop, and in combination with a preset variable precise irrigation decision model, so as to decide the crop irrigation in the soil area of the crop to be decided.

[0138] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Those of ordinary skill in the art, under the inspiration of the present invention, without departing from the spirit of the present invention and the scope protected by the claims, can also make changes, modifications, substitutions, and variations to the above embodiments, and these all fall within the protection scope of the present invention.

[0139] If the technical solution of the present disclosure involves personal information, before the product applying the technical solution of the present disclosure processes personal information, it has clearly informed the personal information processing rules and obtained the individual's independent consent. If the technical solution of the present disclosure involves sensitive personal information, before the product applying the technical solution of the present disclosure processes sensitive personal information, it has obtained the individual's separate consent and at the same time meets the requirements of "express consent". For example, at personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If an individual voluntarily enters the collection scope, it is regarded as consenting to the collection of their personal information; or on the device for personal information processing, when the personal information processing rules are informed by obvious signs / information, personal authorization is obtained through pop-up messages or by asking the individual to upload their personal information by themselves, etc.; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the types of personal information processed.

Claims

1. A crop irrigation decision-making method based on the water demand characteristics of crops during their growth period, characterized in that the steps include: Obtain remote sensing satellite images of the soil area of ​​the crop to be decided and the segmented growth period and transition time attributes of each crop; Monitoring the soil water content of the crop soil area to be decided based on the remote sensing satellite images and the temperature vegetation drought index; According to the soil water content and the segmented growth period and transition time attributes of each crop, combined with a preset variable precision irrigation decision model, a variable irrigation prescription map is generated to decide on crop irrigation in the soil area of ​​the crop to be decided.

2. The crop irrigation decision-making method based on the water demand characteristics of crops during their growth period according to claim 1, characterized in that: The monitoring of soil water content in the crop soil area to be determined includes: Based on the remote sensing satellite images, soil water content, temperature vegetation drought index, surface temperature and normalized difference vegetation index under drought stress are inverted; The soil water content of the crop soil area to be decided is inverted based on the measured soil water content, soil water content under drought stress, temperature vegetation drought index, surface temperature and normalized difference vegetation index.

3. The crop irrigation decision-making method based on the water demand characteristics of crops during their growth period as claimed in claim 2, characterized in that: The variable precision irrigation decision model is obtained by integrating the water demand bidirectional prediction model and the irrigation intensification decision model; the step of generating a variable irrigation prescription map in combination with the preset variable precision irrigation decision model includes: According to the segmented growth period and transition period attributes of each crop and the soil water content, a two-way water demand prediction model is constructed and pre-trained to obtain the water demand and irrigation time interval of the corresponding type of crop; The water demand deviation factor is obtained according to the water tolerance of each growth stage of the corresponding type of crops; According to the water requirement of the corresponding type of crops, meteorological data, soil moisture conditions, irrigation time intervals and water demand deviation factors, the variable irrigation prescription map of the soil area of ​​the crop to be decided is generated through the configured irrigation intensification decision model.

4. The crop irrigation decision-making method based on the water demand characteristics of crops during their growth period as claimed in claim 3, characterized in that: The attributes of each crop segmented growth period and transition period include: corresponding meteorological data during the growth period, evaporation status, soil moisture, crop absorption rate, k The standard water space required for each growth period and the k Meteorological data, evaporation status, soil moisture, and crop absorption rate data during the pre-transition and post-transition periods of each growth period; The steps of constructing and training the water demand bidirectional prediction model include: S301, setting the pre-transition section and post-transition section corresponding to each growth stage of the crop to be decided, and constructing a transition change function according to the historical water demand corresponding to the pre-transition section and post-transition section; The front transition section is a time period corresponding to the growth stage and moving forward by a first preset number of days, and the rear transition section is a time period corresponding to the growth stage and moving backward by a first preset number of days; S302, based on BP neural network and particle swarm algorithm to build a water demand bidirectional prediction model integrated K A water demand prediction sub-model is built, and the transition change function is built into the water demand prediction sub-model corresponding to the adjacent growth period, and the crop water demand dynamic balance prediction equation of the corresponding growth period is built into the water demand prediction sub-model of the corresponding growth period; S303, k The corresponding meteorological data, evaporation status, soil moisture, crop absorption rate, k The standard water space required for each growth period and the k The meteorological data, evaporation status, soil moisture, crop absorption rate data and soil water content of the crop soil area to be decided in the first growth period are input into the k The water demand prediction sub-model corresponding to each growth period is trained to obtain K The training error of transition section prediction and the training error of non-transition section corresponding to each water demand prediction sub-model; S304, set the training error threshold, according to the obtained K The transition segment prediction training error and K -1 non-transition segment training error is used to construct a comprehensive training loss function. When the comprehensive training loss function is less than the training error threshold, a trained water demand prediction sub-model is obtained.

5. The crop irrigation decision-making method based on the water demand characteristics of crops during their growth period as claimed in claim 4, characterized in that: The steps of obtaining the transition segment prediction training error and the non-transition segment training error include: S3031, when k If the water demand prediction time point of the kth growth period is in the non-transition period of the current growth period, the water demand prediction sub-model corresponding to the kth growth period is used to predict the water demand at the corresponding time point, and the predicted water demand and the historical real water demand are used to obtain the kth growth period. k Non-transition segment training error; S3032, when k When the water demand forecast time point of the current growth period is in the pre-transition period of the current growth period, k The water demand prediction sub-model corresponding to each growth period is used for pre-prediction, and the k The pre-transitional period of the growth period corresponds to t The water demand forecast at the moment k -1 The water demand prediction sub-model corresponding to the growth period predicts the water demand forecast at the same time.

6. The crop irrigation decision-making method based on the water demand characteristics of crops during their growth period as claimed in claim 5, characterized in that: The step of obtaining the transition segment prediction training error and the non-transition segment training error further includes: S3033, using k The water demand forecast obtained by the water demand forecast sub-model corresponding to the k-th growth period and the water demand forecast obtained by the water demand forecast sub-model corresponding to the k-1th growth period at the same time point are taken as the average of the water demand forecast obtained by the water demand forecast sub-model corresponding to the k-1th growth period. t The water demand forecast corresponding to the time t The water demand forecast corresponding to the moment and the water demand at the corresponding time point are used to obtain the transition segment prediction training error at the corresponding time point of the preceding transition segment; S3034, when k When the water demand forecast time point of the current growth period is in the post-transition period of the current growth period, repeat the S3032 process to obtain the k The water demand prediction sub-model corresponding to each growth period corresponds to the post-transition period t The water demand forecast at the time and the k+ The water demand prediction sub-model corresponding to one growth period corresponds to the water demand prediction amount at the same time point, and S3033 is repeated to obtain the transition segment prediction training error at the time point corresponding to the post-transition segment.

7. The crop irrigation decision-making method based on the water demand characteristics of crops during their growth period as claimed in claim 6, characterized in that: The step of obtaining the water demand deviation factor comprises: S311, configuring water tolerance experiments for crops at different growth stages, and setting a water requirement bidirectional prediction model to predict the water requirement of the crops at the corresponding growth stage as the ideal water requirement; S312, taking the ideal water requirement as the starting point, increasing or decreasing the water requirement of the crop in the corresponding growth period, and monitoring the crop growth status corresponding to the corresponding water requirement at each moment, and automatically evaluating the crop growth status score under the corresponding water requirement using the configured evaluation algorithm; S313, using the water requirement and the growth status score under the corresponding water requirement, obtaining a water requirement-growth score change curve corresponding to different growth stages of the corresponding crop during the entire growth period; S314, taking the crop growth status score corresponding to the ideal water requirement in each growth stage as the starting point, when the water requirement is increased or decreased and the growth status score corresponding to the water requirement is continuous m When the water requirement is less than the crop growth status score corresponding to the ideal water requirement, the minimum water requirement and the maximum water requirement corresponding to the crop growth status score starting to decrease in the corresponding growth period are obtained; S315. According to the minimum water requirement and the ideal water requirement, obtain the lower limit deviation factor of the water requirement of the corresponding growth period, and according to the maximum water requirement and the ideal water requirement, obtain the upper limit deviation factor of the water requirement of the corresponding growth period.

8. The crop irrigation decision-making method based on the water demand characteristics of crops during their growth period as claimed in claim 7, characterized in that: The steps of constructing the irrigation intensification decision model include: S401, constructing a water requirement interval corresponding to each growth stage of the crop based on the ideal water requirement, the lower limit deviation factor of the water requirement and the upper limit deviation factor of the water requirement; S402: Using the water demand interval of each growth stage as a variable irrigation prescription map, and setting irrigation strategy execution action trigger information according to the variable irrigation prescription map.

9. The crop irrigation decision-making method based on the water demand characteristics of crops during their growth period as claimed in claim 8, characterized in that: The step of setting the irrigation strategy execution action trigger information comprises: S4021. When the water holding capacity of the corresponding crop growth area at the current moment is lower than the lower limit of the water demand range in the variable irrigation prescription map, irrigation is directly carried out until the ideal water demand of the corresponding growth stage is reached; S4022, when the corresponding water holding capacity of the corresponding crop growth area at the current moment is lower than the ideal water requirement of the corresponding growth stage, the parameters of the irrigation intensification decision model are pre-adjusted according to the ideal water requirement of the corresponding growth stage; when the water holding capacity of the corresponding crop growth area reaches the lower limit of the water requirement range in the variable irrigation prescription map, the irrigation device is controlled to irrigate by pre-adjusting the parameters of the irrigation intensification decision model until the ideal water requirement of the corresponding growth stage is reached; S4023. When the water holding capacity of the crop growth area at the current moment is higher than the ideal water requirement of the corresponding growth stage, no irrigation is performed. When it is lower than the ideal water requirement of the corresponding growth stage, the process of S4022 is repeated.

10. A crop irrigation decision system based on the water demand characteristics of crops during their growth period, which is used to implement the crop irrigation decision method based on the water demand characteristics of crops during their growth period as claimed in any one of claims 1 to 9, characterized in that: include: Prediction module and irrigation decision module; The prediction module is used to monitor the soil water content of the crop soil area to be decided based on the remote sensing satellite image and the temperature vegetation drought index; The irrigation decision module is used to generate a variable irrigation prescription map based on the soil water content and the segmented growth period and transition time period attributes of each crop, combined with a preset variable precision irrigation decision model, to decide on crop irrigation in the soil area of ​​the crop to be decided.

Images