Regional crop planting management method and apparatus

Abstract

The application provides a regional crop planting management method and device, and applies to the technical field of agricultural production management. The method comprises the following steps: determining a target crop according to a crop growth scene of a target region; performing global sensitivity analysis on a crop growth model based on physiological parameters of the target crop, soil parameters of the target region and management parameters, obtaining sensitive parameters, and performing localized calibration on the crop growth model based on the sensitive parameters; coupling a calibrated crop growth model with a pesticide transport model and a nutrient transport model to construct a comprehensive growth simulation system; combining soil parameters and cost factors of the target region; and traversing and simulating environmental and income changes under different planting strategies through the comprehensive growth simulation system to determine a target crop planting strategy.

Description

Technical Field

[0001] This invention relates to the field of agricultural production management technology, and in particular to a method and apparatus for regional crop planting management. Background Technology

[0002] In recent years, with the development of computer technology, remote sensing technology, and crop growth models, agricultural management has gradually transformed towards precision and scientific management.

[0003] In existing precision planting management schemes, planting strategy generation mainly relies on traditional experience and precise simulations using crop growth models. Crop growth models are widely used to predict crop growth status and guide fertilization and irrigation decisions. These models consider multiple factors such as meteorological conditions, soil type, and crop variety, generating planting management schemes by simulating the crop's fertilizer and water requirements at different growth and development stages. Remote sensing technology provides spatial and temporal monitoring capabilities. Existing schemes often utilize remote sensing imagery to acquire crop growth data, analyze indicators such as crop growth vigor, chlorophyll content, and nitrogen concentration, and then dynamically adjust fertilization amounts. Researchers can assess crop growth status in real time and adjust fertilization strategies promptly based on changes in remote sensing imagery.

[0004] However, existing crop growth models and remote sensing data assimilation technologies still have many shortcomings, including insufficient simulation of environmental impact, limited effectiveness of single algorithms, the balance between computational accuracy and efficiency, and the lack of an integrated closed-loop system for planting management. Summary of the Invention

[0005] This invention provides a regional crop planting management method and apparatus to address the shortcomings of existing crop growth models and remote sensing data assimilation technologies.

[0006] This invention provides a regional crop planting management method, comprising: determining a target crop based on the crop growth scenario of a target region; performing a global sensitivity analysis on a crop growth model based on the physiological parameters of the target crop, the soil parameters of the target region, and management parameters to obtain sensitive parameters, and performing localized calibration of the crop growth model based on the sensitive parameters; constructing a comprehensive growth simulation system by coupling the calibrated crop growth model with a pesticide transport model and a nutrient transport model, and combining the soil parameters and cost factors of the target region, traversing and simulating environmental and profit changes under different planting strategies through the comprehensive growth simulation system to determine the target crop planting strategy.

[0007] According to a regional crop planting management method provided by the present invention, the method involves performing a global sensitivity analysis on a crop growth model based on the physiological parameters of the target crop, the soil parameters of the target region, and management parameters to obtain sensitive parameters, and then performing localized calibration of the crop growth model based on the sensitive parameters. This includes: establishing a multidimensional parameter screening mechanism using the Extended Fourier Transform (EFT) amplitude sensitivity test to perform a global sensitivity analysis on some input parameters of the crop growth model to obtain a global sensitivity index for each parameter; identifying parameters with a global sensitivity index greater than a first threshold as sensitive parameters; and using a data assimilation algorithm to back-calculate the sensitive parameters of the crop growth model based on field-collected parameters to obtain localized crop parameters. The partial input parameters of the crop growth model include the crop parameters of the target crop, the soil parameters of the target region, and the management parameters; the sensitive parameters include sensitive crop parameters, sensitive soil parameters, and sensitive management parameters.

[0008] According to a regional crop planting management method provided by the present invention, the method of constructing a comprehensive growth simulation system by coupling a calibrated crop growth model with a pesticide transport model and a nutrient transport model includes: running the crop growth model to obtain the full-cycle physiological parameters of the target crop, wherein the full-cycle physiological parameters include physiological and biochemical parameters and growth cycle parameters; inputting the full-cycle physiological parameters into the pesticide transport model to obtain pesticide residues; inputting the full-cycle physiological parameters into the nutrient transport model to obtain soil nutrient content; and inputting the soil nutrient content into the crop growth model as the initial condition for the next round of simulation.

[0009] According to a regional crop planting management method provided by the present invention, after determining the target crop planting strategy, the method further includes: during the growth and development stage of the target crop, retrieving a time series of ground parameters based on remote sensing data, wherein the parameter types of the time series of ground parameters include meteorological parameters, leaf area index, soil moisture, and chlorophyll content; assimilating the time series of ground parameters with the crop growth model, and performing inversion calculations on sensitive parameters affecting crop growth based on the assimilated crop growth model to determine the crop nutrient status in the soil; and spatially differentiating the target crop planting strategy by monitoring the crop nutrient status in the soil to obtain a cultivation strategy, wherein the cultivation strategy includes the amount and timing of irrigation, the type and amount of fertilizer, and the timing and dosage of pesticide application.

[0010] According to a regional crop planting management method provided by the present invention, the cultivation strategy is obtained by spatially differentiating the planting strategy of the target crop by monitoring the crop nutrient status in the soil, including: predicting the cultivation strategy by using a long short-term memory neural network model LSTM as input, with the time series of the crop nutrient status as input; wherein the input of the long short-term memory neural network model LSTM is the time series of the crop nutrient status, and the output is a sensitive soil parameter.

[0011] According to a regional crop planting management method provided by the present invention, after determining the target crop planting strategy, the method further includes: using the 4DSA-EnSRF assimilation algorithm to predict the spatial yield distribution of the target crop; wherein, the 4DSA-EnSRF assimilation algorithm integrates simulated annealing algorithm, four-dimensional variational algorithm, ensemble root mean square filter algorithm and variational time window mechanism.

[0012] This invention also provides a regional crop planting management device, comprising the following modules: a crop determination module, a localization calibration module, and a strategy determination module; the crop determination module is used to determine the target crop based on the crop growth scenario of the target area; the localization calibration module is used to perform a global sensitivity analysis on the crop growth model based on the physiological parameters of the target crop, the soil parameters of the target area, and management parameters to obtain sensitive parameters, and to perform localization calibration on the crop growth model based on the sensitive parameters; the strategy determination module is used to couple the calibrated crop growth model with a pesticide transport model and a nutrient transport model to construct a comprehensive growth simulation system, and, in conjunction with the soil parameters and cost factors of the target area, to traverse and simulate environmental and profit changes under different planting strategies through the comprehensive growth simulation system to determine the target crop planting strategy.

[0013] According to a regional crop planting management device provided by the present invention, the localization calibration module is used to establish a multi-dimensional parameter screening mechanism using the extended Fourier amplitude sensitivity test method, perform global sensitivity analysis on some input parameters of the crop growth model, and obtain the global sensitivity index of each parameter; determine the parameters whose global sensitivity index is greater than a first threshold as sensitive parameters; and use a data assimilation algorithm to back-infer the sensitive parameters of the crop growth model based on field-collected parameters to obtain localized crop parameters; wherein, some input parameters of the crop growth model include crop parameters of the target crop, soil parameters of the target area, and management parameters, and the sensitive parameters include sensitive crop parameters, sensitive soil parameters, and sensitive management parameters.

[0014] According to a regional crop planting management device provided by the present invention, the strategy determination module is used to run a crop growth model to obtain the full-cycle physiological parameters of the target crop, the full-cycle physiological parameters including physiological and biochemical parameters and growth cycle parameters; input the full-cycle physiological parameters into the pesticide transport model to obtain pesticide residues, input the full-cycle physiological parameters into the nutrient transport model to obtain soil nutrient content; and input the soil nutrient content into the crop growth model as the initial condition for the next round of simulation.

[0015] According to the present invention, a regional crop planting management device further includes: a fine management module; the fine management module is used to, during the growth and development stage of the target crop, retrieve a time series of ground parameters based on remote sensing data, wherein the parameter types of the time series of ground parameters include meteorological parameters, leaf area index, soil moisture, and chlorophyll content; assimilate the time series of ground parameters with the crop growth model, and perform inversion calculations on sensitive parameters affecting crop growth based on the assimilated crop growth model to determine the crop nutrient status in the soil; and perform spatial differentiation processing on the planting strategy of the target crop by monitoring the crop nutrient status in the soil to obtain a cultivation strategy, wherein the cultivation strategy includes the amount and timing of irrigation, the type and amount of fertilizer, and the timing and dosage of pesticide application.

[0016] According to a regional crop planting management device provided by the present invention, the fine management module is used to predict the cultivation strategy by using a long short-term memory neural network model (LSTM) with the time series of crop nutrient status as input; wherein, the input of the long short-term memory neural network model (LSTM) is the time series of crop nutrient status, and the output is a sensitive soil parameter.

[0017] According to the present invention, a regional crop planting management device further includes: a yield prediction module; the yield prediction module is used to predict the spatial yield distribution of the target crop using a 4DSA-EnSRF assimilation algorithm; wherein the 4DSA-EnSRF assimilation algorithm integrates simulated annealing algorithm, four-dimensional variational algorithm, ensemble root mean square filter algorithm and variational time window mechanism.

[0018] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the regional crop planting and management method as described above.

[0019] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the regional crop planting and management method as described above.

[0020] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the regional crop planting and management method as described above.

[0021] The regional crop planting management method and apparatus provided by this invention can determine the target crop based on the crop growth scenario of the target area, thus identifying the most suitable crop variety for planting in the target area and providing a scientific basis for subsequent planting management strategies. Furthermore, by performing global sensitivity analysis on the crop growth model of the target crop to obtain sensitive parameters, it can accurately identify crop parameters that have a significant impact on crop growth, providing a reliable basis for subsequent localization calibration and making the crop growth model more closely match actual growth conditions. Because the crop growth model of the target crop can be locally calibrated based on sensitive parameters, the accuracy and reliability of the crop growth model can be improved, accurately simulating the crop growth process. Since a comprehensive growth simulation system can be constructed, crop growth can be comprehensively considered from multiple dimensions, fully reflecting the complexities of the planting process. Finally, by traversing and simulating environmental and yield changes under different planting strategies through the comprehensive growth simulation system to determine the target crop planting strategy, the advantages and disadvantages of different strategies can be quantified, providing data support for selecting the optimal planting strategy. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1 This is one of the flowcharts illustrating the regional crop planting and management method provided by the present invention;

[0024] Figure 2 This is a schematic diagram of the coupling framework of the integrated growth simulation system provided by the present invention;

[0025] Figure 3 This is the second flowchart of the regional crop planting management method provided by the present invention;

[0026] Figure 4 This is the third flowchart of the regional crop planting and management method provided by the present invention;

[0027] Figure 5 This is a flowchart illustrating the data assimilation algorithm provided by the present invention;

[0028] Figure 6 This is a schematic diagram of the structure of the regional crop planting management device provided by the present invention;

[0029] Figure 7 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0031] It should be noted that in the embodiments of this application, the words "exemplary" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0032] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0033] To facilitate a clear description of the technical solutions of the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish the same or similar items with essentially the same function and effect. Those skilled in the art can understand that the terms "first" and "second" are not intended to limit the quantity or execution order.

[0034] This application describes some exemplary embodiments for illustrative purposes. It should be understood that this application may be implemented in other ways not specifically shown in the accompanying drawings.

[0035] like Figure 1 As shown, this application provides a regional crop planting management method, which can be applied to a regional crop planting management device. The regional crop planting management method may include steps S101-S103:

[0036] S101, The regional crop planting management device determines the target crop based on the crop growth scenario of the target area.

[0037] Specifically, firstly, the regional crop planting management device acquires data related to crop growth scenarios in the target area, such as meteorological parameters, soil parameters, and economic data. Meteorological parameters can be daily data (minimum / maximum temperature, rainfall, solar radiation, etc.) from the past 5 years and predicted for the next 3 years. Soil parameters can be stratified measurements of mechanical composition, field capacity (FC), available nitrogen (SAN), available phosphorus (SAP), available potassium (SAK), wilting point, etc. Economic data can include crop market price fluctuations, mechanized planting costs, and government subsidy policies. Then, based on the acquired data, the regional crop planting management device conducts growth simulation experiments on different crops to determine the most suitable target crop for the target area.

[0038] Alternatively, growth simulation experiments for different crops can be conducted using crop models such as WOFOST and DSSAT.

[0039] S102. The regional crop planting management device performs a global sensitivity analysis on the crop growth model based on the physiological parameters of the target crop, the soil parameters of the target area, and the management parameters to obtain sensitive parameters, and performs localized calibration of the crop growth model based on the sensitive parameters.

[0040] Specifically, regional crop planting management devices can first determine a crop growth model that matches the target crop. For example, if the target crop is sugarcane, the crop growth model can be the WOFOST-Sugarcane (WS) model. The WS model mainly simulates changes in crop environmental water balance, sugarcane canopy development, and biomass of various components under various management conditions. The WS model can simulate sugarcane growth under various management conditions and is widely used in climate change, different sugarcane planting areas, and different management methods.

[0041] Optionally, the regional crop planting management device performs a global sensitivity analysis on the crop growth model based on the physiological parameters of the target crop, the soil parameters of the target area, and the management parameters to obtain sensitive parameters, and performs localized calibration of the crop growth model based on the sensitive parameters. This includes: establishing a multidimensional parameter screening mechanism using the extended Fourier amplitude sensitivity test to perform a global sensitivity analysis on some input parameters of the crop growth model to obtain a global sensitivity index for each parameter; identifying parameters with a global sensitivity index greater than a first threshold as sensitive parameters; and using a data assimilation algorithm to back-calculate the sensitive parameters of the crop growth model based on field-collected parameters to obtain localized crop parameters. The partial input parameters of the crop growth model include the crop parameters of the target crop, the soil parameters of the target area, and the management parameters, and the sensitive parameters include sensitive crop parameters, sensitive soil parameters, and sensitive management parameters.

[0042] Specifically, since crop growth models have numerous input parameters, sensitivity analysis methods are needed to identify sensitive parameters to reduce workload. For example, the Extended Fourier Amplitude Sensitivity Test (EFAST) can be used to establish a multidimensional parameter screening mechanism. This allows for global sensitivity analysis of some input parameters of the crop growth model to obtain the global sensitivity index for each parameter. Then, based on the ranking of sensitivity indices, a parameter classification standard is established, defining parameters with sensitivity indices above a first threshold as sensitive parameters and the rest as insensitive parameters.

[0043] It should be noted that the crop parameters for the target crop can include leaf area index (LAI), stem weight (SM), leaf weight (LM), and aboveground biomass (BIO). The soil parameters for the target area can include porosity, saturated hydraulic conductivity, etc., while the management parameters can include planting / harvesting dates, irrigation / fertilization amounts, and corresponding times.

[0044] Optionally, for sensitive parameters, the regional crop planting management device can use the Particle Swarm Optimization (PSO) algorithm for dynamic calibration; for non-sensitive parameters, the regional crop planting management device can directly call the soil survey dataset (soil parameters) and the model default values ​​(management and crop parameters).

[0045] Specifically, the regional crop planting management device performs localized calibration of the crop growth model based on the sensitive parameters, including: obtaining crop parameters for the entire growth period of the target crop through multi-gradient field trials, continuously adjusting the sensitive parameters using the PSO algorithm, narrowing the gap between the simulation results of the crop growth model and the parameters collected in the field, thereby obtaining the sensitive crop parameters that best match the growth and development of the target crop in the target area.

[0046] S103. The regional crop planting management device uses a calibrated crop growth model coupled with a pesticide transport model and a nutrient transport model to construct a comprehensive growth simulation system. Combining the soil parameters and cost factors of the target area, the comprehensive growth simulation system traverses and simulates the environmental and profit changes under different planting strategies to determine the target crop planting strategy.

[0047] For detailed issues such as pesticide residues and nitrogen loss in soil, since crop growth models are not as accurate as specialized pesticide or nitrogen migration models, crop growth models can be coupled with pesticide transport models and nutrient transport models to construct a comprehensive growth simulation system, so as to achieve data sharing and functional complementarity.

[0048] Optionally, the pesticide transport model can be a pesticide root zone model (PRZM), and the nutrient transport model can be a nitrate leaching and economic analysis package (NLEAP).

[0049] The PRZM model input includes seven categories of parameters: meteorological parameters (daily precipitation, wind speed, temperature, solar radiation), soil parameters (soil bulk density, porosity, organic carbon content, etc.), water body parameters (water body characteristics such as volume, depth, and surface area; rate constants for processes such as biodegradation, hydrolysis, and photolysis in water; partition coefficients of chemical substances between water and soil), pesticide chemical parameters (pesticide molecular weight, vapor pressure, solubility, Henry's law constant, and degradation half-life in soil, water, and plants, etc.), pesticide management parameters (including application methods of chemical substances (such as ground application, foliar spraying, etc.), application date, and application rate), crop physiological parameters (maximum root depth, maximum canopy cover, maximum canopy interception, etc.), and crop growth cycle parameters (crop emergence date, maturity date, harvest date, etc.). Some input parameters (meteorological parameters, soil parameters) and output parameters (crop growth cycle parameters, crop physiological parameters) of the crop growth model can be used as input parameters for the PRZM model. The inclusion of the crop growth model can enhance its simulation effect while reducing the workload of parameter investigation for the PRZM model.

[0050] The NLEAP model input includes four types of parameters: meteorological parameters (daily precipitation, wind speed, temperature, solar radiation, daily evaporation, etc.), soil parameters (soil depth, organic matter content, total nitrogen content, available nitrogen content, field capacity, etc.), management parameters (crop type, planting date, harvest date, nitrogen fertilizer application rate, fertilization time, fertilization method, irrigation time, irrigation amount, etc.), and crop parameters (variety name, maximum root depth, carbon-nitrogen ratio of harvested portion, carbon-nitrogen ratio of unharvested portion, maximum yield, etc.). Most of the meteorological, soil, and management parameters overlap with crop growth models. Crop parameters can also be provided by the simulation results of crop growth models. Including crop growth models can enhance the simulation effect while reducing the workload of NLEAP model parameter surveys.

[0051] Optionally, the regional crop planting management device utilizes a calibrated crop growth model coupled with a pesticide transport model and a nutrient transport model to construct a comprehensive growth simulation system, including: running the crop growth model to obtain the full-cycle physiological parameters of the target crop, wherein the full-cycle physiological parameters include physiological and biochemical parameters and growth cycle parameters; inputting the full-cycle physiological parameters into the pesticide transport model to obtain pesticide residues; inputting the full-cycle physiological parameters into the nutrient transport model to obtain soil nutrient content; and inputting the soil nutrient content into the crop growth model as the initial condition for the next round of simulation.

[0052] Specifically, such as Figure 2As shown, the process of constructing a comprehensive growth simulation system for regional crop planting and management devices includes: first, running a crop growth model, inputting meteorological parameters, crop parameters, soil parameters, and management parameters into the model to simulate growth and obtain crop physiological parameters and growth cycle. Then, inputting the crop growth model's input parameters, crop physiological parameters, growth cycle, along with pesticide chemical parameters, water parameters, and pesticide management parameters, into the PRZM model to obtain pesticide residue information such as soil pesticide distribution, water pesticide distribution, crop pesticide residues, and pesticide degradation and transformation. Finally, inputting the crop growth model's input parameters, crop physiological parameters, growth cycle, along with other meteorological parameters, other crop parameters, other soil parameters, and other management parameters into the NLEAP model to obtain soil nitrogen content, i.e., soil nutrient content, such as nitrate soil distribution, nitrate leaching, nitrogen loss, and denitrification. This method provides a more accurate representation of crop growth and development, making the simulation of environmental impacts by both models more accurate. Based on the results of the PRZM and NLEAP models, pesticide and nitrate residues in the target area can be obtained, thereby assessing the carrying capacity of the current environment and determining whether to continue the current planting strategy. Finally, the soil nutrient content output by the NLEAP model can be used as a soil parameter input into the crop growth model as the initial condition for the next round of simulation.

[0053] After coupling the three models, the yield data of the target crop under different management modes can be obtained by combining the historical meteorological data and local soil data of the target area. On this basis, combined with the soil conditions and recent meteorological conditions of the target area, and taking into account the prices of irrigation, fertilizer, pesticides and labor, various management strategies are digitized. Then, the environmental changes and benefit changes under different management strategies are simulated in an ergonomic manner. Based on the principle of maximizing total benefit, the target crop planting strategy is obtained. The target crop planting strategy includes, but is not limited to, common agricultural planting management measures such as planting date, planting density, fertilizer amount and fertilization date, irrigation amount and irrigation date, pesticide amount and pesticide application date, crop rotation and rotation method.

[0054] Optionally, the total revenue of the target crop is calculated as follows:

[0055] ;

[0056] in, This represents the total revenue from planting the target crop. Indicates the number of years since planting. Indicates the first Annual target crop planting revenue The discount rate represents the time value of money. Indicates the first The cost of ecological restoration after the new year.

[0057] ;

[0058] in, Indicates the first Annual unit price This represents the crop growth model simulated under a certain management mode. Annual output This indicates the first under this management model Annual policy subsidies, This indicates the cost of seeds, fertilizer, labor, etc. under this management model. This indicates the regulatory compliance costs under this management model, including environmental taxes and carbon emission fees. This indicates environmental costs, including soil remediation and fines for over-extraction of water resources.

[0059] In this embodiment, since the target crop can be determined based on the crop growth scenario of the target area, the most suitable crop variety for planting in the target area can be identified, providing a scientific basis for subsequent planting management strategies. Since a global sensitivity analysis of the crop growth model of the target crop can be performed to obtain sensitive parameters, crop parameters that have a significant impact on crop growth can be accurately identified, providing a reliable basis for subsequent localization calibration and making the crop growth model more closely match actual growth conditions. Since the crop growth model of the target crop can be locally calibrated based on sensitive parameters, the accuracy and reliability of the crop growth model can be improved, accurately simulating the crop growth process. Since a comprehensive growth simulation system can be constructed, crop growth can be comprehensively considered from multiple dimensions, fully reflecting the complexities of the planting process. Since the comprehensive growth simulation system can traverse and simulate environmental and yield changes under different planting strategies to determine the target crop planting strategy, the advantages and disadvantages of different strategies can be quantified, providing data support for selecting the optimal planting strategy.

[0060] like Figure 3 As shown, after determining the target crop planting strategy, the regional crop planting management method provided in this application embodiment may further include S104-S106:

[0061] S104. The regional crop planting management device retrieves the time series of ground parameters based on remote sensing data during the growth and development stages of the target crop.

[0062] The types of parameters in the ground parameter time series include meteorological parameters, leaf area index, soil moisture, and chlorophyll content.

[0063] After determining the target crop planting strategy, the regional crop planting management device can conduct refined management of the target area based on remote sensing data.

[0064] Optionally, Sentinel 1 and Sentinel 2 satellite remote sensing data can be selected. The downloaded remote sensing image data is used to establish correlations with the surface LAI and SM, and the surface LAI and SM are retrieved. Specifically, the LAI is retrieved from Sentinel 2 visible light remote sensing imagery, and the SM is retrieved from Sentinel 1 microwave remote sensing imagery, thus obtaining the time series of ground parameters.

[0065] Optionally, if the target area has insufficient visible light remote sensing imagery due to weather factors, cloud point pixels appearing in some of the remote sensing images can be corrected using interpolation and filtering methods. For example, for discontinuous cloud point pixels, a method of first linear interpolation followed by Savitzky-Golay filtering can be used for correction. For continuous cloud point pixels, the following algorithm can be used for correction:

[0066] ;

[0067] in, LAI represents remote sensing inversion. This represents the LAI obtained from crop growth model simulation. This represents the mean value obtained by removing noise from the remote sensing image. , , , This is an empirical coefficient, which can be determined based on the actual situation.

[0068] S105. The regional crop planting management device assimilates the ground parameter time series with the crop growth model, and performs inversion calculation on the sensitive parameters affecting crop growth based on the assimilated crop growth model to determine the crop nutrient status in the soil.

[0069] Regional crop planting management devices can use the SA algorithm as a data assimilation algorithm to continuously change the values ​​of sensitive parameters, thereby narrowing the gap between remote sensing inversion of LAI and SM and model simulation of LAI and SM, and thus obtaining the distribution of crop nutrient status in the field soil.

[0070] S106. The regional crop planting management device performs spatial differentiation processing on the target crop planting strategy by monitoring the crop nutrient status in the soil to obtain the cultivation strategy.

[0071] The cultivation strategy includes the amount and timing of irrigation, the type and amount of fertilizer, and the timing and dosage of pesticide application.

[0072] Regional crop planting management devices can improve soil conditions in the field based on the crop nutrient status of the soil, enabling precise irrigation, fertilization, and other operations.

[0073] For field fertilization, the amount of fertilizer applied to each plot can be:

[0074] ;

[0075] in, This indicates the amount of fertilizer applied to the target plot (which can be nitrogen, potassium, or phosphorus fertilizer). This indicates the content of available nutrients in the soil (which can be available nitrogen, available phosphorus, and available potassium). This indicates the average content of available nutrients in the soil. This indicates the amount of fertilizer (which can be nitrogen, potassium, or phosphorus) applied according to the target crop planting strategy.

[0076] For field irrigation, the irrigation amount for each plot unit can be:

[0077] ;

[0078] in, The irrigation amount for the target plot unit. Field holding capacity This indicates the soil moisture content of the target plot unit. Irrigation volume determined for the target crop planting strategy. This represents the average field capacity.

[0079] It should be noted that, compared with traditional irrigation strategies, the irrigation strategy provided in this application takes into account both field water holding capacity and soil moisture content, that is, it takes into account the effective water use range of crops. Therefore, the irrigation strategy provided in this application will be more reasonable in the allocation of irrigation water.

[0080] It should be noted that although crop growth models cannot simulate the disease conditions of the target crop, they can monitor the type, severity, and distribution of diseases in the target crop through remote sensing data and disease identification models (such as threshold segmentation and watershed algorithms). This information can then guide pesticide application: applying different types of pesticides to different disease types; promptly replacing planted crops in severely diseased areas to mitigate damage; applying pesticides to less diseased areas to reduce losses; and applying preventative pesticides to areas spatially close to disease-affected areas to prevent disease spread. Furthermore, updating the crop growth model's status in real time based on surface conditions can improve the accuracy of the simulation, further enhancing the precision of predicting the target crop's maturity date, and ultimately guiding harvesting operations.

[0081] Optionally, the regional crop planting management device performs spatial differentiation processing on the target crop planting strategy by monitoring the crop nutrient status in the soil to obtain a cultivation strategy, including: using the time series of the crop nutrient status as input, predicting the cultivation strategy through a Long Short-Term Memory (LSTM) neural network model, wherein the input of the LSTM neural network model is the time series of the crop nutrient status, and the output is a sensitive soil parameter.

[0082] Specifically, the regional crop planting management device can use an LSTM-GRU hybrid architecture to predict the nutrient status in the soil, that is, combining the temporal feature capture capability of bidirectional LSTM with the gating simplification mechanism of Gated Recurrent Unit (GRU) to reduce the redundant computation of the prediction process.

[0083] (1) The architecture design of LSTM is as follows:

[0084] The input layer includes time series of parameters such as surface parameters and meteorological parameters:

[0085] Ground parameters: Time series of leaf area index (LAI) and soil moisture content (SM) during the key growth stages (budding stage, tillering stage, and elongation stage) of the target crop;

[0086] Meteorological parameters: daily data on maximum temperature, minimum temperature, humidity, wind speed, rainfall, and radiation;

[0087] Dimension requirements: Time step ≥ 90 days, input matrix dimension 90×8 (8 feature channels).

[0088] The hidden layer includes:

[0089] A 3-layer bidirectional LSTM (64 neurons per layer) is followed by a GRU unit (32 neurons) after each layer.

[0090] Regularization modules: Dropout layer (deactivation rate 0.2) and Batch Normalization layer.

[0091] The output layer includes:

[0092] The fully connected layer is mapped to the sensitive soil parameters FC, SAN, SAK, and SAP, and the output is a 1×4 matrix.

[0093] (2) The training and optimization of LSTM are as follows:

[0094] Loss function:

[0095] The joint loss function L balances the absolute error and the relative error: L = 0.7RMSE + 0.3MAPE;

[0096] The root mean squared error (RMSE) is an indicator that measures the difference between the predicted value and the actual value, while the mean absolute percentage error (MAPE) represents the average absolute percentage difference between the predicted value and the actual value.

[0097] Optimization strategy:

[0098] Optimizer: Adam (learning rate 0.001, decay rate β1=0.9, β2=0.999);

[0099] Batch size: 32;

[0100] Training cycles: 50 (early stopping method to monitor validation set loss, tolerance 10 cycles).

[0101] (3) Data preparation and verification:

[0102] Dataset Construction:

[0103] Training samples are generated based on the data assimilation results. The inputs are LAI, SM and meteorological time series data, and the output is the assimilated soil parameters. At the same time, the data is divided into training set, validation set and test set according to time order.

[0104] Verification method:

[0105] Time series cross-validation: A 5-fold sliding window (window step size of 15 days) is used to avoid future information leakage;

[0106] This application utilizes the input and output of a data assimilation algorithm to construct a training dataset. The inputs include LAI and SM data from the growing season, as well as time series data of meteorological elements. The outputs are the model assimilation results FC, SAN, SAK, and SAP. The dataset is divided into a training set and a test set. The training set is used for model training, and the test set is used to verify the model's generalization ability. The verification process employs 5-fold time series cross-validation. The system deployment utilizes NVIDIA A100 GPUs for parallel computing, which significantly reduces the single-sample inversion time.

[0107] In this embodiment, since the time series of ground parameters can be retrieved based on remote sensing data and assimilated with the crop growth model, the fusion of measured data and the model can be achieved. This allows the model to better reflect the actual growth situation, thereby more accurately determining the crop nutrient status in the soil and improving the accuracy of model prediction and analysis. Furthermore, since the sensitive parameters affecting crop growth can be retrieved and calculated based on the assimilated crop growth model, and cultivation strategies can be determined by monitoring the crop nutrient status in the soil, the amount and timing of irrigation, the type and amount of fertilizer, and the timing and dosage of pesticide application can be precisely planned. This avoids blind irrigation, fertilization, and pesticide application, preventing resource waste and environmental pollution while meeting crop growth needs, thereby improving resource utilization efficiency.

[0108] like Figure 4 As shown, after determining the target crop planting strategy, the regional crop planting management method provided in this application embodiment may further include S107:

[0109] S107. The regional crop planting management device uses the 4DSA-EnSRF assimilation algorithm to predict the spatial yield distribution of the target crop.

[0110] The 4DSA-EnSRF assimilation algorithm integrates Simulated Annealing (SA), 4-Dimensional Variational Data Assimilation (4DVAR), Ensemble Square Root Filter (EnSRF), and variational time window mechanism.

[0111] Specifically, such as Figure 5 As shown, the implementation of the 4DSA-EnSRF assimilation algorithm mainly consists of the following three steps:

[0112] (1) Minimize the objective function constructed by 4DVAR through SA to optimize the background field sensitive parameters.

[0113] The basic idea of ​​simulated annealing is to perform global optimization by simulating the physical annealing process. Its goal is to minimize the objective function by continuously optimizing the background field state. The update formula for simulated annealing can be expressed as:

[0114] ;

[0115] in, This is the current state. This is the state for the next iteration. The temperature parameter determines the step size. It is a random perturbation from the current state to the next state. This represents the number of iteration steps.

[0116] The annealing process involves gradually decreasing the temperature to control the magnitude of the perturbation. In the algorithm, the objective function for optimizing the state is changed from one that only considers the differences between LAI and SM to one that incorporates 4DVAR. J 4DVAR The expression for the objective function is:

[0117]

[0118] in, This represents a vector of sensitive parameters, which may be model parameters such as "maximum photosynthetic efficiency" or "root depth". This is the background state, i.e., the calibration value of the sensitive parameter. The background error covariance matrix is ​​constructed based on prior knowledge, historical data, or experience; t represents time t. This represents the observation vector, i.e., the specific observed value; It is the observation matrix, representing the simulated values ​​from the model; It is the observation error covariance matrix, which represents the variance of each observation value from the true value and the covariance between different observation values. It is also constructed based on prior knowledge, historical data or experience. This objective function integrates time, space, model and observation information. Compared with the original function, this function is more effective and interpretable.

[0119] (2) Calculate Kalman gain using EnSRF algorithm

[0120] First, through the steps described above, a set of optimized background field state samples (i.e., the background set) is predicted. Each sample represents a possible state of the system at the current moment. Assuming the system state is x, and by assuming a background error of ±5%, a background set can be obtained: , where N is the number of samples.

[0121] Calculate the mean and covariance of the background set:

[0122] Mean: ;

[0123] Covariance: ;

[0124] Then, the state of the background set is mapped to the observation space using the observation matrix H of the remote sensing inversion data. That is, based on the state of each sample, the corresponding observation value is calculated:

[0125] ;

[0126] Kalman gain calculation:

[0127] ;

[0128] in, Indicates Kalman gain, Represents the background error covariance matrix. This represents the transpose of the observation matrix H. This represents the observation error covariance matrix, which numerically represents the deviation between the observed value and the true value, and is determined based on historical research or literature experience.

[0129] (3) Introduce variational time window to adjust Kalman gain

[0130] The LAI (Local Area Index) is higher during the crop's vigorous growth stage, making observations more reliable. However, in the early stages, due to lower LAI, the LAI retrieved from remote sensing has more noise and lower reliability. In the later stages, when leaves turn yellow, reliability is even lower. Therefore, a beta distribution can be used to construct the weighting function. Assuming the target crop's growing period is one year (365 days), the weights can be set as follows:

[0131] ;

[0132] ;

[0133] ;

[0134] in, The weight is represented by t, where t∈[1,365] represents the number of days after planting. For the normalized time variable, Represents the beta distribution, where the beta distribution is set. =5, making The weight is lower when it is close to 0; set =3, making It decays more slowly as it approaches 1, but not as low as it did in the early stages.

[0135] according to Adjust the Kalman gain, and then update the analysis set. Assume the original Kalman gain is... The Kalman gain is updated as follows:

[0136] ;

[0137] Update analysis set as follows:

[0138] ;

[0139] Finally, by updating each sample in the background set, a new analysis set is obtained. This analysis set represents the state estimate corrected based on the observation data. Using the state estimate to update the state predictions of the crop growth model can optimize the model's estimate of the target crop yield.

[0140] The 4DSA-EnSRF assimilation algorithm aims to enhance the model's constraint on spatiotemporal evolution by globally optimizing the background state and combining it with a four-dimensional variational assimilation objective function. Furthermore, a variational time window mechanism is introduced to dynamically weight the Kalman gain, thereby improving the simulation accuracy of crop growth states during the assimilation process. This algorithm effectively addresses issues such as inconsistencies between observation errors and temporal resolution, and is particularly suitable for remote sensing-driven regional-scale crop simulation scenarios.

[0141] In this embodiment, the 4DSA-EnSRF assimilation algorithm can be used to predict the yield of the target crop under the target crop planting strategy. Since the 4DSA-EnSRF assimilation algorithm integrates simulated annealing, four-dimensional variational algorithm, ensemble root mean square filtering, and variational time window mechanism, it can more effectively address the error accumulation problems caused by inconsistent input data time steps and large initial background field deviations, thereby improving the accuracy and interpretability of crop growth simulation. This method is particularly suitable for agricultural production environments with high uncertainty and can significantly improve the yield prediction capability of the target crop under specific planting strategies.

[0142] The foregoing mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, it includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the units and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0143] The regional crop planting management method provided in this application can be implemented by a regional crop planting management device or a control module for regional crop planting management within that device. This application uses the execution of the regional crop planting management method by a regional crop planting management device as an example to illustrate the regional crop planting management device provided in this application.

[0144] It should be noted that the embodiments of this application can divide the regional crop planting management device into functional modules according to the above method examples. For example, each function can be divided into its own functional modules, or two or more functions can be integrated into one processing module. The integrated modules can be implemented in hardware or as software functional modules. Optionally, the module division in the embodiments of this application is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0145] like Figure 6 As shown in the figure, this application embodiment provides a regional crop planting management device 600. The regional crop planting management device 600 includes: a crop determination module 601, a localization calibration module 602, and a strategy determination module 603. The crop determination module 601 is used to determine the target crop based on the crop growth scenario of the target area. The localization calibration module 602 is used to perform a global sensitivity analysis on the crop growth model based on the physiological parameters of the target crop, the soil parameters of the target area, and management parameters to obtain sensitive parameters, and to perform localization calibration on the crop growth model based on the sensitive parameters. The strategy determination module 603 is used to couple the calibrated crop growth model with a pesticide transport model and a nutrient transport model to construct a comprehensive growth simulation system. Combining the soil parameters and cost factors of the target area, the comprehensive growth simulation system traverses and simulates environmental and profit changes under different planting strategies to determine the target crop planting strategy.

[0146] Optionally, the localization calibration module 602 is used to establish a multi-dimensional parameter screening mechanism using the extended Fourier amplitude sensitivity test method, perform global sensitivity analysis on some input parameters of the crop growth model, and obtain the global sensitivity index of each parameter; determine the parameters whose global sensitivity index is greater than a first threshold as sensitive parameters; and use a data assimilation algorithm to back-infer the sensitive parameters of the crop growth model based on the field-collected parameters to obtain localized crop parameters; wherein, some input parameters of the crop growth model include crop parameters of the target crop, soil parameters of the target area, and management parameters, and the sensitive parameters include sensitive crop parameters, sensitive soil parameters, and sensitive management parameters.

[0147] Optionally, the strategy determination module 603 is used to run a crop growth model to obtain the full-cycle physiological parameters of the target crop, the full-cycle physiological parameters including physiological and biochemical parameters and growth cycle parameters; input the full-cycle physiological parameters into the pesticide transport model to obtain pesticide residues, input the full-cycle physiological parameters into the nutrient transport model to obtain soil nutrient content; and input the soil nutrient content into the crop growth model as the initial condition for the next round of simulation.

[0148] Optionally, the crop planting management device 600 in this area further includes: a fine management module 604; the fine management module 604 is used to retrieve the time series of ground parameters based on remote sensing data during the growth and development stage of the target crop, wherein the parameter types of the time series of ground parameters include meteorological parameters, leaf area index, soil moisture, and chlorophyll content; to assimilate the time series of ground parameters with the crop growth model, and to perform inversion calculations on the sensitive parameters affecting crop growth based on the assimilated crop growth model to determine the crop nutrient status in the soil; and to perform spatial differentiation processing on the planting strategy of the target crop by monitoring the crop nutrient status in the soil to obtain a cultivation strategy, wherein the cultivation strategy includes the amount and timing of irrigation, the type and amount of fertilizer, and the timing and dosage of pesticide application.

[0149] Optionally, the fine management module 604 is used to predict the cultivation strategy using a long short-term memory neural network model LSTM as input, with the time series of the crop nutrient status as input; wherein the input of the long short-term memory neural network model LSTM is the time series of the crop nutrient status, and the output is a sensitive soil parameter.

[0150] Optionally, the crop planting management device 600 in this area further includes: a yield prediction module 605; the yield prediction module 605 is used to predict the spatial yield distribution of the target crop using the 4DSA-EnSRF assimilation algorithm; wherein the 4DSA-EnSRF assimilation algorithm integrates simulated annealing algorithm, four-dimensional variational algorithm, ensemble root mean square filter algorithm and variational time window mechanism.

[0151] In this embodiment, since the target crop can be determined based on the crop growth scenario of the target area, the most suitable crop variety for planting in the target area can be identified, providing a scientific basis for subsequent planting management strategies. Since a global sensitivity analysis of the crop growth model of the target crop can be performed to obtain sensitive parameters, crop parameters that have a significant impact on crop growth can be accurately identified, providing a reliable basis for subsequent localization calibration and making the crop growth model more closely match actual growth conditions. Since the crop growth model of the target crop can be locally calibrated based on sensitive parameters, the accuracy and reliability of the crop growth model can be improved, accurately simulating the crop growth process. Since a comprehensive growth simulation system can be constructed, crop growth can be comprehensively considered from multiple dimensions, fully reflecting the complexities of the planting process. Since the comprehensive growth simulation system can traverse and simulate environmental and yield changes under different planting strategies to determine the target crop planting strategy, the advantages and disadvantages of different strategies can be quantified, providing data support for selecting the optimal planting strategy.

[0152] Figure 7An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 7 As shown, the electronic device may include a processor 710, a communications interface 720, a memory 730, and a communication bus 740, wherein the processor 710, communications interface 720, and memory 730 communicate with each other via the communication bus 740. The processor 710 can call logical instructions in the memory 730 to execute a regional crop planting management method. This method includes: determining a target crop based on the crop growth scenario of the target region; performing a global sensitivity analysis on a crop growth model based on the physiological parameters of the target crop, the soil parameters of the target region, and management parameters to obtain sensitive parameters, and performing localized calibration of the crop growth model based on the sensitive parameters; using the calibrated crop growth model coupled with a pesticide transport model and a nutrient transport model to construct a comprehensive growth simulation system; combining the soil parameters and cost factors of the target region, and using the comprehensive growth simulation system to traverse and simulate environmental and profit changes under different planting strategies to determine the target crop planting strategy.

[0153] Furthermore, the logical instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0154] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the regional crop planting management method provided by the above methods. The method includes: determining the target crop based on the crop growth scenario of the target region; performing a global sensitivity analysis on the crop growth model based on the physiological parameters of the target crop, the soil parameters of the target region, and management parameters to obtain sensitive parameters, and performing localized calibration of the crop growth model based on the sensitive parameters; using the calibrated crop growth model coupled with a pesticide transport model and a nutrient transport model to construct a comprehensive growth simulation system, and combining the soil parameters and cost factors of the target region, using the comprehensive growth simulation system to traverse and simulate environmental and profit changes under different planting strategies to determine the target crop planting strategy.

[0155] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the regional crop planting management method provided by the above methods. This method includes: determining a target crop based on the crop growth scenario of the target region; performing a global sensitivity analysis on a crop growth model based on the physiological parameters of the target crop, the soil parameters of the target region, and management parameters to obtain sensitive parameters, and performing localized calibration of the crop growth model based on the sensitive parameters; constructing a comprehensive growth simulation system by coupling the calibrated crop growth model with a pesticide transport model and a nutrient transport model; and, in conjunction with the soil parameters and cost factors of the target region, traversing and simulating environmental and profit changes under different planting strategies through the comprehensive growth simulation system to determine the target crop planting strategy.

[0156] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0157] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0158] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for regional crop planting and management, characterized in that, include: Determine the target crop based on the crop growth scenario in the target area; A global sensitivity analysis of the crop growth model is performed based on the physiological parameters of the target crop, the soil parameters of the target area, and the management parameters to obtain sensitive parameters. The crop growth model is then calibrated locally based on the sensitive parameters. A comprehensive growth simulation system is constructed by coupling a calibrated crop growth model with a pesticide transport model and a nutrient transport model. Combining the soil parameters and cost factors of the target area, the comprehensive growth simulation system is used to simulate the environmental and profit changes under different planting strategies in order to determine the target crop planting strategy. The process involves performing a global sensitivity analysis on the crop growth model based on the physiological parameters of the target crop, the soil parameters of the target region, and management parameters to obtain sensitive parameters, and then performing localized calibration of the crop growth model based on these sensitive parameters, including: A multidimensional parameter screening mechanism is established using the Extended Fourier Amplitude Sensitivity Test (EFT) to perform global sensitivity analysis on some input parameters of the crop growth model, obtaining the global sensitivity index for each parameter. Parameters with a global sensitivity index greater than a first threshold are identified as sensitive parameters. Using a data assimilation algorithm, the sensitive parameters of the crop growth model are inferred from field-collected parameters to obtain localized crop parameters. The input parameters of the crop growth model include crop parameters of the target crop, soil parameters of the target area, and management parameters. The sensitive parameters include sensitive crop parameters, sensitive soil parameters, and sensitive management parameters. The integrated growth simulation system constructed by coupling a calibrated crop growth model with pesticide transport and nutrient transport models includes: The crop growth model is run to obtain the full-cycle physiological parameters of the target crop, which include physiological and biochemical parameters and growth cycle parameters; the full-cycle physiological parameters are input into the pesticide transport model to obtain pesticide residues, and the full-cycle physiological parameters are input into the nutrient transport model to obtain soil nutrient content; the soil nutrient content is input into the crop growth model as the initial condition for the next round of simulation. After determining the target crop planting strategy, the method further includes: During the growth and development stages of the target crop, time series of ground parameters are retrieved based on remote sensing data. The types of parameters in the time series include meteorological parameters, leaf area index, soil moisture, and chlorophyll content. The time series of ground parameters are assimilated with the crop growth model. Based on the assimilated crop growth model, sensitive parameters affecting crop growth are retrieved and calculated to determine the crop nutrient status in the soil. By monitoring the crop nutrient status in the soil, the planting strategy for the target crop is spatially differentiated to obtain a cultivation strategy. The cultivation strategy includes the amount and timing of irrigation, the type and amount of fertilizer, and the timing and dosage of pesticide application.

2. The regional crop planting and management method according to claim 1, characterized in that, The cultivation strategy is obtained by spatially differentiating the planting strategy of the target crop by monitoring the crop nutrient status in the soil, including: Using the time series of crop nutrient status as input, the cultivation strategy is predicted through a long short-term memory neural network model (LSTM). The input to the Long Short-Term Memory (LSTM) neural network model is a time series of crop nutrient status, and the output is a sensitive soil parameter.

3. The regional crop planting and management method according to claim 1, characterized in that, After determining the target crop planting strategy, the method further includes: The spatial yield distribution of the target crop was predicted using the 4DSA-EnSRF assimilation algorithm. The 4DSA-EnSRF assimilation algorithm integrates simulated annealing, four-dimensional variational algorithm, ensemble root mean square filter algorithm, and variational time window mechanism.

4. A regional crop planting and management device, characterized in that, include: Crop determination module, localization calibration module, strategy determination module; The crop determination module is used to determine the target crop based on the crop growth scenario of the target area; The localization calibration module is used to perform a global sensitivity analysis on the crop growth model based on the physiological parameters of the target crop, the soil parameters of the target area, and the management parameters to obtain sensitive parameters, and to perform localization calibration on the crop growth model based on the sensitive parameters. The process of performing a global sensitivity analysis on the crop growth model based on the physiological parameters of the target crop, the soil parameters of the target area, and the management parameters to obtain sensitive parameters, and then performing localization calibration on the crop growth model based on the sensitive parameters, includes: establishing a multi-dimensional parameter screening mechanism using the extended Fourier amplitude sensitivity test method; performing a global sensitivity analysis on some input parameters of the crop growth model to obtain a global sensitivity index for each parameter; identifying parameters with a global sensitivity index greater than a first threshold as sensitive parameters; and using a data assimilation algorithm to back-calculate the sensitive parameters of the crop growth model based on field-collected parameters to obtain localized crop parameters. The partial input parameters of the crop growth model include the crop parameters of the target crop, the soil parameters of the target area, and the management parameters; the sensitive parameters include sensitive crop parameters, sensitive soil parameters, and sensitive management parameters. The strategy determination module is used to construct a comprehensive growth simulation system by coupling a calibrated crop growth model with pesticide transport and nutrient transport models. Combining soil parameters and cost factors of the target area, the comprehensive growth simulation system iterates through environmental and profit changes under different planting strategies to determine the target crop planting strategy. The construction of the comprehensive growth simulation system by coupling the calibrated crop growth model with pesticide transport and nutrient transport models includes: running the crop growth model to obtain the full-cycle physiological parameters of the target crop, including physiological and biochemical parameters and growth cycle parameters; inputting the full-cycle physiological parameters into the pesticide transport model to obtain pesticide residues; inputting the full-cycle physiological parameters into the nutrient transport model to obtain soil nutrient content; and inputting the soil nutrient content into the crop growth model as the initial condition for the next round of simulation. After determining the target crop planting strategy, the process further includes: during the growth and development stages of the target crop, retrieving time series of ground parameters based on remote sensing data. The types of parameters in the time series include meteorological parameters, leaf area index, soil moisture, and chlorophyll content. The time series of ground parameters is then assimilated with the crop growth model. Based on the assimilated crop growth model, sensitive parameters affecting crop growth are inverted and calculated to determine the crop nutrient status in the soil. The target crop planting strategy is then spatially differentiated by monitoring the crop nutrient status in the soil to obtain a cultivation strategy. This cultivation strategy includes the amount and timing of irrigation, the type and amount of fertilizer applied, and the timing and dosage of pesticide application.

5. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the regional crop planting management method as described in any one of claims 1 to 3.

6. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the regional crop planting management method as described in any one of claims 1 to 3.

7. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the regional crop planting management method as described in any one of claims 1 to 3.

Images