An intelligent agricultural planting management method and system

Abstract

This invention discloses an intelligent agricultural planting management method and system, belonging to the field of smart agriculture technology. The method includes data layer construction, water and fertilizer prediction layer construction, planting optimization layer construction, and agricultural planting management. The invention employs a spatiotemporally adaptive water and fertilizer prediction layer construction method that combines crop physiological load with crop growth stages. This method accurately predicts the water requirements and nitrogen, phosphorus, and potassium nutrient needs of crops in future stages, dynamically responding to crop growth status and environmental changes. It takes into account the differences in water and fertilizer requirements at different growth stages and improves the accuracy and applicability of predictions through spatiotemporal fusion, thus facilitating precision fertilization. The invention also employs a planting optimization layer construction method that combines digital twins and multi-objective optimization. Considering crop growth stages, soil nutrient status, and future weather conditions, this method refines and dynamically optimizes water and fertilizer management strategies, generating optimal management strategies that increase yield, conserve water and fertilizer resources, and reduce the risk of nitrogen leaching.

Description

Technical Field

[0001] This invention belongs to the field of smart agriculture technology, specifically referring to an intelligent agricultural planting management method and system. Background Technology

[0002] Intelligent agricultural planting management utilizes multi-source agricultural data and intelligent decision-making technology to manage crops throughout the entire process, from monitoring and forecasting to water and fertilizer regulation. This enables dynamic control of agricultural production, aiming to improve crop yield and quality, enhance water and fertilizer resource utilization efficiency, and reduce environmental risks, thereby promoting the development of agricultural production towards precision, efficiency, and green practices.

[0003] However, in the existing intelligent agricultural planting management process, there are technical problems such as the inability of traditional water and fertilizer management methods to accurately reflect the actual physiological needs of crops and ignore the differences in water and nutrient requirements at different growth stages, resulting in uneven water and fertilizer application, resource waste and unstable crop yield and quality; there are also technical problems such as the inability of traditional water and fertilizer management strategies to simultaneously take into account crop yield, resource utilization efficiency and environmental protection, and the inability to cope with future weather uncertainties and dynamic changes in crop growth status. Summary of the Invention

[0004] To address the above issues and overcome the shortcomings of existing technologies, this invention provides an intelligent agricultural planting management method and system. It creatively employs a water and fertilizer prediction layer construction method that combines crop physiological load with crop growth stages, enabling precise prediction of future crop water requirements and nitrogen, phosphorus, and potassium nutrient needs. This method dynamically responds to crop growth status and environmental changes, taking into account the differences in water and fertilizer requirements at different growth stages. Furthermore, it enhances the accuracy and applicability of predictions through spatiotemporal fusion, thereby facilitating precision fertilization. The invention also creatively employs a planting optimization layer construction method that combines digital twins and multi-objective optimization. This method enables refined and dynamic optimization of water and fertilizer management strategies, considering crop growth stages, soil nutrient status, and future weather conditions. This results in the generation of optimal, executable management strategies that increase yield, conserve water and fertilizer resources, and reduce the risk of nitrogen leaching.

[0005] The technical solution adopted by this invention is as follows: This invention provides an intelligent agricultural planting management method, which includes the following steps:

[0006] Step S1: Data layer construction;

[0007] Step S2: Construction of the water and fertilizer prediction layer;

[0008] Step S3: Construction of the planting optimization layer;

[0009] Step S4: Agricultural planting management.

[0010] Furthermore, in step S1, the data layer construction specifically involves collecting multi-channel remote sensing image data, field environment monitoring data, future weather forecasts, and crop growth information, and performing preprocessing, spatiotemporal alignment, and standardization to obtain a spatiotemporal fusion dataset.

[0011] Further, in step S2, the water and fertilizer prediction layer is constructed to predict the water and nutrient requirements of crops. Specifically, based on a spatiotemporal fusion dataset, a water and fertilizer prediction layer construction method combining crop physiological load and crop growth stage is used to obtain water and fertilizer demand prediction information, including the following steps:

[0012] Step S21: Input sequence construction, specifically by calculating crop physiological load factors and combining multi-channel remote sensing image data, environmental time series data and future weather forecasts to construct spatial input sequences, temporal input sequences and weather forecast sequences;

[0013] The crop physiological load factor is used to characterize the combined degree of water stress, heat stress and evapotranspiration pressure on crops under current environmental conditions.

[0014] Step S22: Parallel feature extraction, used to extract spatial and temporal features in parallel. Specifically, a dual-branch feature extraction network is constructed to extract spatial and temporal features in parallel. The dual-branch feature extraction network includes a spatial branch and a temporal branch. The spatial branch extracts spatial features by performing convolutional feature extraction on the spatial input sequence through a lightweight convolutional neural network. The temporal branch captures the changing trends of the temporal input sequence through a two-layer recurrent neural network to obtain the temporal hidden state. The temporal hidden state is then concatenated with the weather forecast sequence and mapped through a fully connected layer to obtain the temporal features.

[0015] Step S23: Spatiotemporal consistency fusion, used to generate globally spatiotemporally consistent fusion features. Specifically, by calculating the consistency between spatial features and temporal features, a spatiotemporal adaptability coefficient is constructed. Then, the context vector formed by concatenating the one-hot encoding of crop growth stage and crop physiological load factor is combined, and dynamic weights are generated through a gating network. Adaptive weight correction is performed based on the spatiotemporal adaptability coefficient. Finally, the spatial features and temporal features are weighted and fused to obtain the spatiotemporal fusion features.

[0016] Step S24: Multi-objective prediction output, specifically, multi-output regression prediction is performed on the spatiotemporal fusion features through a multi-task learning model, and a comprehensive loss function is constructed by introducing agricultural proportion constraints and spatiotemporal consistency constraints to correct the prediction results and obtain water and fertilizer demand prediction information.

[0017] The agricultural ratio constraint is used to ensure that the predicted values ​​of nitrogen, phosphorus, and potassium meet the agricultural experience ratio relationship.

[0018] The spatiotemporal consistency constraint is used to maintain the consistency between spatial features and temporal features in the spatiotemporal fusion features.

[0019] Further, in step S3, the planting optimization layer is constructed to generate an optimal water and fertilizer management strategy that takes into account yield, resource, and environmental benefits. Specifically, based on water and fertilizer prediction information and a spatiotemporal fusion dataset, a planting optimization layer construction method combining digital twins and multi-objective optimization is used to obtain the optimal water and fertilizer management strategy, including the following steps:

[0020] Step S31: Initialize the decision space, which is used to construct an optimal decision space for controlling water and fertilizer in the presence of weather forecast errors. Specifically, it involves constructing water and fertilizer decision vectors, setting dynamic search intervals, introducing a set of weather scenarios to construct a multi-objective optimization function, setting multiple constraints, and obtaining a multi-objective decision space model.

[0021] The water and fertilizer decision vector includes irrigation amount and nitrogen, phosphorus and potassium application amount;

[0022] The aforementioned setting of a dynamic search range constructs a feasible search range centered on the predicted value and adaptively adjusted according to the reproductive stage;

[0023] The multi-objective optimization function is used to simultaneously evaluate yield improvement, water resource utilization efficiency and nitrogen leaching risk in the presence of weather forecast errors. It includes a negative yield objective function, a water consumption objective function, and a nitrogen leaching risk objective function.

[0024] The aforementioned setting of multiple constraints specifically involves constraining soil moisture content, nutrient ratio, and equipment capacity.

[0025] Step S32: Digital twin simulation, used to simulate crop growth process based on candidate water and fertilizer management strategies. Specifically, by constructing a digital twin including a crop growth mechanism model, a data proxy model and a physical consistency correction module, the system receives field environmental monitoring data, future weather forecasts and candidate water and fertilizer management strategies as inputs, performs time-series simulation of yield and nutrient residue, and performs constraint correction through the physical consistency correction module to obtain crop growth simulation results.

[0026] The crop growth mechanism model is used to simulate water transport, evapotranspiration, changes in soil moisture content, and potential crop growth states.

[0027] The data proxy model is specifically a multi-output regression model based on machine learning, used for preliminary prediction of yield and soil nutrient residue.

[0028] Step S33: Multi-objective optimization solution, used to obtain stable and executable water and fertilizer management strategies under multiple constraints and weather forecast errors. Specifically, it combines the decision space initialization and the digital twin inference, and introduces a dynamic penalty mechanism, a robust fitness mechanism and a constraint-first two-layer selection mechanism on the basis of the multi-objective evolutionary algorithm to adaptively optimize the candidate water and fertilizer management strategies and obtain a set of candidate water and fertilizer management strategies.

[0029] The dynamic penalty mechanism specifically employs an exponential penalty factor that increases with the number of iterations to gradually suppress constraint violations.

[0030] The robust fitness mechanism constructs a robust fitness by introducing a set of weather scenarios, and evaluates the robustness of candidate water and fertilizer management strategies under multiple weather scenarios.

[0031] The constraint-first two-layer selection mechanism specifically involves prioritizing the selection of water and fertilizer management strategies that fully satisfy multiple constraints during the non-dominated sorting process, and then adaptively sorting and selecting candidate solutions based on the comprehensive fitness to achieve iterative evolution of the candidate water and fertilizer management strategy set.

[0032] Step S34: Preference interaction decision, which is used to generate the final water and fertilizer management strategy from the strategy set based on user preferences. Specifically, by obtaining the user's preference weights for yield, water conservation and environmental risk, the candidate water and fertilizer management strategy set is weighted and re-sorted to select the optimal water and fertilizer management strategy that meets the preference constraints.

[0033] Furthermore, in step S4, the agricultural planting management specifically involves integrating the data layer, the water and fertilizer prediction layer, and the planting optimization layer by executing steps S1 to S3, constructing a water and fertilizer optimization model for agricultural planting management, optimizing water and fertilizer, generating field-level control instructions based on the optimal water and fertilizer management strategy, sending them to the integrated water and fertilizer control system, monitoring the execution effect, and generating a planting management feedback report.

[0034] The present invention provides an intelligent agricultural planting management system, including a data establishment module, a water and fertilizer prediction module, a planting optimization module, and an agricultural planting management module;

[0035] The data establishment module is used for data layer construction. Through data layer construction, a spatiotemporal fusion dataset is obtained, and the spatiotemporal fusion dataset is sent to the water and fertilizer prediction module and the planting optimization module.

[0036] The water and fertilizer prediction module is used to construct the water and fertilizer prediction layer. Through the construction of the water and fertilizer prediction layer, water and fertilizer demand prediction information is obtained, and the water and fertilizer demand prediction information is sent to the planting optimization module.

[0037] The planting optimization module is used to construct a planting optimization layer. Through the construction of the planting optimization layer, the optimal water and fertilizer management strategy is obtained, and the optimal water and fertilizer management strategy is sent to the agricultural planting management module.

[0038] The agricultural planting management module is used for agricultural planting management, and obtains field-level control instructions and planting management feedback reports.

[0039] The beneficial effects achieved by the present invention using the above solution are as follows:

[0040] (1) In the existing intelligent agricultural planting management process, traditional water and fertilizer management methods are difficult to accurately reflect the actual physiological needs of crops and ignore the differences in water and nutrient requirements at different growth stages, resulting in uneven water and fertilizer application, resource waste and unstable crop yield and quality. This solution creatively adopts a water and fertilizer prediction layer construction method that combines crop physiological load and crop growth stage, which realizes accurate prediction of the water requirement and nitrogen, phosphorus and potassium nutrient requirements of crops in the future stage. It can dynamically respond to crop growth status and environmental changes, take into account the differences in water and fertilizer requirements at different growth stages, and improve the accuracy and applicability of prediction through spatiotemporal integration, thereby helping to achieve precision fertilization.

[0041] (2) In response to the technical problems that traditional water and fertilizer management strategies in the existing intelligent agricultural planting management process are difficult to simultaneously take into account crop yield, resource utilization efficiency and environmental protection, and are difficult to cope with future weather uncertainties and dynamic changes in crop growth status, this solution creatively adopts a planting optimization layer construction method that combines digital twins and multi-objective optimization. It realizes the refined and dynamic optimization of water and fertilizer management strategies under the consideration of crop growth stage, soil nutrient status and future weather conditions, thereby generating the optimal executable management strategy that can improve yield, save water and fertilizer resources and reduce nitrogen leaching risk. Attached Figure Description

[0042] Figure 1 A flowchart illustrating an intelligent agricultural planting management method provided by the present invention;

[0043] Figure 2 A schematic diagram of a module of an intelligent agricultural planting management system provided by the present invention;

[0044] Figure 3 This is a flowchart illustrating step S2;

[0045] Figure 4 This is a flowchart illustrating step S3.

[0046] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation

[0047] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0048] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0049] Example 1, see Figure 1 The present invention provides an intelligent agricultural planting management method, which includes the following steps:

[0050] Step S1: Data layer construction;

[0051] Step S2: Construction of the water and fertilizer prediction layer;

[0052] Step S3: Construction of the planting optimization layer;

[0053] Step S4: Agricultural planting management.

[0054] Example 2, see Figure 1 This embodiment is based on the above embodiment. In step S1, the data layer is constructed to provide a unified and standardized sensing data base for the water and fertilizer prediction layer and the planting optimization layer. Specifically, it is obtained by collecting multi-channel remote sensing image data, field environment monitoring data, future weather forecasts and crop growth information, and performing preprocessing, spatiotemporal alignment and standardization processing to obtain a spatiotemporal fusion dataset.

[0055] The multi-channel remote sensing image data includes the Normalized Difference Vegetation Index (NDVI), the Red Edge Vegetation Index (REDI), and the Crop Water Stress Index (CMS); the field environment monitoring data includes soil moisture content, soil electrical conductivity, air temperature, air humidity, and solar radiation intensity over the past 7 days; the future weather forecast includes rainfall predictions, reference evapotranspiration predictions, and temperature predictions for the next 24, 48, and 72 hours; and the crop growth information includes crop variety, crop growth stage, and planting density.

[0056] The preprocessing includes data cleaning, anomaly removal, and missing value repair.

[0057] Example 3, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S2, the water and fertilizer prediction layer is constructed to predict the water and nutrient requirements of crops. Specifically, based on the spatiotemporal fusion dataset, a water and fertilizer prediction layer construction method combining crop physiological load and crop growth stage is adopted to obtain water and fertilizer requirement prediction information, including the following steps:

[0058] Step S21: Input sequence construction, specifically by calculating crop physiological load factors and combining multi-channel remote sensing image data, environmental time series data and future weather forecasts to construct spatial input sequences, temporal input sequences and weather forecast sequences;

[0059] The crop physiological load factor is used to characterize the combined degree of water stress, heat stress, and evapotranspiration pressure on crops under current environmental conditions. The calculation formula is as follows:

[0060] ;

[0061] In the formula, PSI is the crop physiological load factor. Crop Water Stress Weights (CWSI) is the Crop Water Stress Index, used to characterize the degree of water shortage in crops. It is the heat stress weight, used to modulate the degree of influence of crop temperature stress, Tem air It's the temperature, Tem opt It is the optimal temperature for crops. It is the evapotranspiration pressure weight, and VPD is the saturated vapor pressure difference, which is used to characterize the degree of atmospheric dryness and the evapotranspiration potential of crops;

[0062] The spatial input sequence consists of multi-channel remote sensing image snapshots composed of normalized vegetation index, red-edge vegetation index, crop water stress index and crop physiological load factor.

[0063] The time input sequence consists of an environmental change sequence comprising soil moisture content, soil electrical conductivity, air temperature, air humidity, solar radiation intensity, and crop physiological load factors over the past 7 days.

[0064] The weather forecast sequence includes rainfall forecasts, reference evapotranspiration forecasts, and temperature forecasts for the next 24, 48, and 72 hours.

[0065] Step S22: Parallel feature extraction, used to extract spatial and temporal features in parallel. Specifically, a dual-branch feature extraction network is constructed to extract spatial and temporal features in parallel. The dual-branch feature extraction network includes a spatial branch and a temporal branch. The spatial branch extracts spatial features by performing convolutional feature extraction on the spatial input sequence through a lightweight convolutional neural network. The temporal branch captures the changing trends of the temporal input sequence through a two-layer recurrent neural network to obtain the temporal hidden state. The temporal hidden state is then concatenated with the weather forecast sequence and mapped through a fully connected layer to obtain the temporal features.

[0066] Preferably, the lightweight convolutional neural network adopts the EfficientNet-B0 architecture, which includes 5 convolutional layers and 3 fully connected layers. The convolutional kernel size is 3×3. After each convolution, the ReLU activation function is used, and the output features are mapped through the fully connected layers.

[0067] Preferably, the two-layer recurrent neural network is a two-layer LSTM with 128 hidden units per layer and a random deactivation rate of 0.2.

[0068] The formula for calculating the timing branch is:

[0069] ;

[0070] ;

[0071] In the formula, h t This is the temporal hidden state at the last time step of the two-layer LSTM, where T is the total time step length, LSTM(·) is the two-layer LSTM operation function, and A is the hidden state at the last time step. temporal It is a time input sequence, V temporal It is a temporal characteristic, w f These are mapping weights, and [·||·] are concatenation operators. A futre It is a weather forecast sequence, b f It is a mapping bias term;

[0072] Step S23: Spatiotemporal consistency fusion, used to generate globally spatiotemporally consistent fusion features. Specifically, by calculating the consistency between spatial features and temporal features, a spatiotemporal adaptability coefficient is constructed. Then, the context vector formed by concatenating the one-hot encoding of crop growth stage and crop physiological load factor is combined, and dynamic weights are generated through a gating network. Adaptive weight correction is performed based on the spatiotemporal adaptability coefficient. Finally, the spatial features and temporal features are weighted and fused to obtain the spatiotemporal fusion features.

[0073] The formula for calculating the spatiotemporal adaptability coefficient is as follows:

[0074] ;

[0075] In the formula, V is the spatiotemporal fitness coefficient, exp(·) is the natural exponential function, and ||·||2 is the L2 norm, used to measure the degree of difference between spatial and temporal features. spatial It is a spatial feature;

[0076] The dynamic weights include spatial dynamic weights and temporal dynamic weights, which are used to adjust the relative importance of spatial features and temporal features in the fusion process. The calculation formula is as follows:

[0077] ;

[0078] ;

[0079] In the formula, C is the context vector, and Stage is the stage vector. OneHot It is a unique thermal code for crop growth stages. It is a spatial dynamic weight, and Sig(·) is the Sigmoid activation function. It is the first weight of gating. It is the first bias term of the gating. It is the second weight of the gating. It is a time-series dynamic weight. It is the second bias term of the gating;

[0080] The formula for calculating the adaptive weight correction is as follows:

[0081] ;

[0082] In the formula, It is the corrected spatial dynamic weight. It is the corrected time-series dynamic weight;

[0083] The calculation formula for weighted fusion of spatial features and temporal features is as follows:

[0084] ;

[0085] In the formula, V fused It is a feature of spatiotemporal fusion;

[0086] Step S24: Multi-objective prediction output, used to obtain the prediction results of water and fertilizer demand in the next 24 to 72 hours. Specifically, a multi-task learning model is used to perform multi-output regression prediction on the spatiotemporal fusion features, and an agricultural proportion constraint and spatiotemporal consistency constraint are introduced to construct a comprehensive loss function to correct the prediction results and obtain water and fertilizer demand prediction information. The water and fertilizer demand prediction information includes the crop water demand and nitrogen, phosphorus and potassium nutrient demand in the next 24 to 72 hours.

[0087] The multi-task learning model adopts a structure of shared fully connected coding layer and multiple prediction branches. First, the spatiotemporal fusion features are learned as a whole through the shared fully connected coding layer to extract shared features. Then, the shared features are input into independent water demand prediction branches, nitrogen prediction branches, phosphorus prediction branches and potassium prediction branches to perform multi-output regression prediction and output non-negative constraint prediction results. Each branch consists of a fully connected layer with ReLU activation function and a linear output layer.

[0088] The aforementioned agricultural ratio constraint is used to ensure that the predicted values ​​of nitrogen, phosphorus, and potassium meet the empirical ratio relationships in agriculture, and to prevent the prediction results from deviating from actual agricultural management standards. The calculation formula is as follows:

[0089] ;

[0090] In the formula, L ratio It is an agricultural proportion constraint. It is a proportional constraint weight. This is the predicted nitrogen nutrient requirement. This is the predicted phosphorus nutrient requirement. It is a small constant used to avoid the case where the denominator is zero, r NP It is an empirical ratio of nitrogen and phosphorus. This is the predicted potassium nutrient requirement, r NK It is an empirical ratio of nitrogen to potassium;

[0091] The spatiotemporal consistency constraint is used to maintain the consistency between spatial features and temporal features in the spatiotemporal fusion features, and to reduce the impact of feature conflicts on the prediction results. The calculation formula is as follows:

[0092] ;

[0093] In the formula, These are consistency constraint weights;

[0094] The formula for calculating the comprehensive loss function is as follows:

[0095] ;

[0096] In the formula, L final It is the comprehensive loss function, L MSE It is a multi-objective mean squared error loss function.

[0097] By performing the above operations, this solution addresses the technical problems in existing intelligent agricultural planting management, where traditional water and fertilizer management methods fail to accurately reflect the actual physiological needs of crops and ignore the differences in water and nutrient requirements at different growth stages, leading to uneven water and fertilizer application, resource waste, and unstable crop yield and quality. This solution creatively adopts a water and fertilizer prediction layer construction method that combines crop physiological load with crop growth stages. This achieves accurate prediction of the crop's future water requirements and nitrogen, phosphorus, and potassium nutrient needs. It can dynamically respond to crop growth status and environmental changes, taking into account the differences in water and fertilizer requirements at different growth stages. Furthermore, it improves the accuracy and applicability of predictions through spatiotemporal fusion, thereby contributing to precision fertilization.

[0098] Example 4, see Figure 1 and Figure 4 This embodiment is based on the above embodiment. In step S3, the planting optimization layer is constructed to generate an optimal water and fertilizer management strategy that takes into account yield, resource and environmental benefits. Specifically, based on water and fertilizer prediction information and spatiotemporal fusion dataset, a planting optimization layer construction method combining digital twin and multi-objective optimization is used to obtain the optimal water and fertilizer management strategy, including the following steps:

[0099] Step S31: Initialize the decision space, which is used to construct an optimal decision space for controlling water and fertilizer in the presence of weather forecast errors. Specifically, it involves constructing water and fertilizer decision vectors, setting dynamic search intervals, introducing a set of weather scenarios to construct a multi-objective optimization function, setting multiple constraints, and obtaining a multi-objective decision space model.

[0100] The water and fertilizer decision vector includes irrigation amount and nitrogen, phosphorus, and potassium application rates, and the calculation formula is as follows:

[0101] ;

[0102] In the formula, X i W is the water and fertilizer decision vector for field i, representing the water and fertilizer management strategy for field i, where i is the field index and W is the field index. i N is the irrigation amount for field i. i It is the nitrogen application rate in field i, P i The phosphorus application rate (K) for field i i This refers to the potassium application rate in field i;

[0103] The dynamic search interval is set to construct a feasible search range centered on the predicted value and adaptively adjusted according to the reproductive stage. The calculation formula is as follows:

[0104] ;

[0105] In the formula, X ij It is the j-th water and fertilizer decision component of field i. It is a search width coefficient that dynamically changes with the crop's growth stage, with a value range of [0.1, 0.3]. It is the j-th water and fertilizer prediction component of field i, which is specifically obtained from the water and fertilizer demand prediction information;

[0106] The multi-objective optimization function is used to simultaneously evaluate yield improvement, water resource utilization efficiency and nitrogen leaching risk in the presence of weather forecast errors. It includes a negative yield objective function, a water consumption objective function, and a nitrogen leaching risk objective function.

[0107] The formula for calculating the negative output objective function is as follows:

[0108] ;

[0109] In the formula, f1(X) is the negative yield objective function, X is the water and fertilizer management strategy, which is composed of the water and fertilizer decision vectors of all fields, and E[·] is the mathematical expectation operation. It is a random weather scenario, following a probability distribution constructed based on future weather forecasts and their errors. Water and fertilizer management strategies under stochastic weather scenarios Production forecasts below;

[0110] The formula for calculating the water consumption objective function is as follows:

[0111] ;

[0112] In the formula, f2(X) is the objective function for water usage, and WaterUsage(X) is the water usage corresponding to the water and fertilizer management strategy;

[0113] The formula for calculating the nitrogen leaching risk objective function is as follows:

[0114] ;

[0115] In the formula, f3(X) is the objective function for nitrogen leaching risk. Water and fertilizer management strategies under stochastic weather scenarios Nitrogen leaching loss at the following levels It is the risk aversion coefficient, and Var[·] is the variance operation;

[0116] The formula for calculating nitrogen leaching loss is as follows:

[0117] ;

[0118] In the formula, It is an empirical constant. This is a predicted value for nitrogen residue; ET is based on stochastic weather scenarios. Evapotranspiration under random weather scenarios, Rain Rainfall;

[0119] The aforementioned setting of multiple constraints specifically involves imposing constraints on soil moisture content, nutrient ratios, and equipment capacity to ensure the agronomic feasibility of the strategy. The calculation formula is as follows:

[0120] ;

[0121] In the formula, It refers to soil moisture content under water and fertilizer management strategies. It is the moisture content at the wilting point, r min It is the lower limit of the nutrient application ratio, r max It is the upper limit of the nutrient application ratio, W max This is the maximum water supply capacity of the irrigation system;

[0122] Step S32: Digital twin simulation, used to simulate crop growth process based on candidate water and fertilizer management strategies. Specifically, by constructing a digital twin including a crop growth mechanism model, a data proxy model and a physical consistency correction module, the system receives field environmental monitoring data, future weather forecasts and candidate water and fertilizer management strategies as inputs, performs time-series simulation of yield and nutrient residue, and performs constraint correction through the physical consistency correction module to obtain crop growth simulation results.

[0123] The crop growth mechanism model is used to simulate water transport, evapotranspiration, soil moisture content changes and potential crop growth states. Specifically, it is a process simulation model based on the crop, soil and atmospheric continuum, which can be implemented using AquaCrop-OS, SWAP or DSSAT models.

[0124] The data proxy model is specifically a multi-output regression model based on machine learning, used for preliminary prediction of yield and soil nutrient residue. The calculation formula is as follows:

[0125] ;

[0126] In the formula, This is the production forecast, f ML (·) represents the data proxy model, which can specifically employ a multilayer perceptron model, A. temporal It is a time input sequence, A futre It is a weather forecast sequence;

[0127] The physical consistency correction module is used to ensure that the projected yield meets the crop physiological mechanism or water balance law, and the calculation formula is as follows:

[0128] ;

[0129] In the formula, Y is the corrected production forecast. It is the soil moisture content correction weight. It is the current soil moisture content predicted by the crop growth mechanism model. It is a reference value for soil moisture content under ideal conditions in crop growth mechanism models. It is the crop growth correction weight, K c It is the current crop growth coefficient derived from the crop growth mechanism model. It is a reference value for crop growth coefficient under ideal conditions in the crop growth mechanism model;

[0130] Step S33: Multi-objective optimization solution, used to obtain stable and executable water and fertilizer management strategies under multiple constraints and weather forecast errors. Specifically, it combines the decision space initialization and the digital twin inference, and introduces a dynamic penalty mechanism, a robust fitness mechanism and a constraint-first two-layer selection mechanism on the basis of the multi-objective evolutionary algorithm to adaptively optimize the candidate water and fertilizer management strategies and obtain a set of candidate water and fertilizer management strategies.

[0131] The dynamic penalty mechanism specifically employs an exponential penalty factor that increases with the number of iterations to gradually suppress constraint violations. The calculation formula is as follows:

[0132] ;

[0133] In the formula, Penalty(t) is an exponential penalty factor. It is the penalty coefficient. Here, t is the growth rate, k is the number of iterations, k is the constraint index, max(·) is the maximum value function, and g is the maximum value function. k (X) is the violation amount of the k-th constraint;

[0134] The robust fitness mechanism constructs a robust fitness score by introducing a set of weather scenarios, and evaluates the robustness of candidate water and fertilizer management strategies under multiple weather scenarios. The calculation formula is as follows:

[0135] ;

[0136] In the formula, F robust (X) is the robustness fitness, m is the objective function index, M is the number of objective functions, and w m It is the weight of the m-th objective function, f m (X) is the m-th objective function. It is an uncertainty-sensitive parameter. It represents the variance of negative output under different weather scenarios, used to measure robustness;

[0137] The constraint-first two-layer selection mechanism specifically involves prioritizing the selection of water and fertilizer management strategies that fully satisfy multiple constraints during the non-dominated sorting process, and then adaptively sorting and selecting candidate solutions based on the comprehensive fitness to achieve iterative evolution of the candidate water and fertilizer management strategy set.

[0138] The formula for calculating the overall fitness is:

[0139] ;

[0140] In the formula, F total (X) represents the overall fitness;

[0141] Step S34: Preference Interaction Decision, used to generate the final water and fertilizer management strategy from the strategy set based on user preferences. Specifically, by obtaining the user's preference weights for yield, water conservation, and environmental risk, the candidate water and fertilizer management strategy set is weighted and re-ranked to select the optimal water and fertilizer management strategy that satisfies the preference constraints. The calculation formula is as follows:

[0142] ;

[0143] ;

[0144] In the formula, p represents the preference weight, p1 is the preference weight for the first objective function (specifically, the negative yield objective function), p2 is the preference weight for the second objective function (specifically, the water consumption objective function), and p3 is the preference weight for the third objective function (specifically, the nitrogen leaching risk objective function). This is the optimal water and fertilizer management strategy. Selected from a pool of candidate water and fertilizer management strategies Minimize the water and fertilizer management strategy X, where PaSet is the set of candidate water and fertilizer management strategies, p m The m-th objective function preference weight, f m (X) is the m-th objective function.

[0145] By performing the above operations, this solution addresses the technical challenges in existing intelligent agricultural planting management processes. Traditional water and fertilizer management strategies struggle to simultaneously consider crop yield, resource utilization efficiency, and environmental protection, and are ill-equipped to cope with future weather uncertainties and dynamic changes in crop growth status. This solution creatively employs a planting optimization layer construction method that combines digital twins and multi-objective optimization. This method enables refined and dynamic optimization of water and fertilizer management strategies, taking into account crop growth stages, soil nutrient status, and future weather conditions. As a result, it generates an optimal and executable management strategy that can both increase yield, conserve water and fertilizer resources, and reduce the risk of nitrogen leaching.

[0146] Example 5, see Figure 1This embodiment is based on the above embodiment. In step S4, the agricultural planting management specifically involves integrating the data layer, the water and fertilizer prediction layer and the planting optimization layer by executing steps S1 to S3, constructing a water and fertilizer optimization model for agricultural planting management, optimizing water and fertilizer, generating field-level control instructions based on the optimal water and fertilizer management strategy, sending them to the integrated water and fertilizer control system, monitoring the execution effect, and generating a planting management feedback report.

[0147] Example 6, see Figure 2 Based on the above embodiments, this embodiment provides an intelligent agricultural planting management system, including a data establishment module, a water and fertilizer prediction module, a planting optimization module, and an agricultural planting management module.

[0148] The data establishment module is used for data layer construction. Through data layer construction, a spatiotemporal fusion dataset is obtained, and the spatiotemporal fusion dataset is sent to the water and fertilizer prediction module and the planting optimization module.

[0149] The water and fertilizer prediction module is used to construct the water and fertilizer prediction layer. Through the construction of the water and fertilizer prediction layer, water and fertilizer demand prediction information is obtained, and the water and fertilizer demand prediction information is sent to the planting optimization module.

[0150] The planting optimization module is used to construct a planting optimization layer. Through the construction of the planting optimization layer, the optimal water and fertilizer management strategy is obtained, and the optimal water and fertilizer management strategy is sent to the agricultural planting management module.

[0151] The agricultural planting management module is used for agricultural planting management, and obtains field-level control instructions and planting management feedback reports.

[0152] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, 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 process, method, article, or apparatus.

[0153] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

[0154] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.

Claims

1. An intelligent agricultural planting management method, characterized in that: The method comprises the following steps: Step S1: data layer construction; Step S2: water and fertilizer prediction layer construction, for predicting water and nutrient requirements of crops, specifically, according to the spatio-temporal fusion data set, a water and fertilizer prediction layer construction method combining crop physiological load and crop growth stage is adopted to obtain water and fertilizer demand prediction information, comprising the following steps: step S21 input sequence construction, step S22 parallel feature extraction, step S23 spatio-temporal consistency fusion and step S24 multi-target prediction output; In step S21, the input sequence construction, specifically, by calculating the crop physiological load factor, combining multi-channel remote sensing image data, environmental time series data and future weather forecast, constructing spatial input sequence, time input sequence and weather forecast sequence; the crop physiological load factor is used to represent the comprehensive degree of water stress, heat stress and evapotranspiration pressure of crops under current environmental conditions; In step S22, the parallel feature extraction, specifically, by constructing a double-branch feature extraction network, parallelly extracting spatial features and time series features, the double-branch feature extraction network comprises a spatial branch and a time series branch; the spatial branch extracts convolutional features from the spatial input sequence through a lightweight convolutional neural network to obtain spatial features; the time series branch captures the trend of the time input sequence through a double-layer recurrent neural network to obtain a time series hidden state, and splices the time series hidden state with the weather forecast sequence, and then maps through a fully connected layer to obtain a time series feature; In step S23, the spatio-temporal consistency fusion is used to generate globally spatio-temporally consistent fusion features, specifically, by calculating the consistency of the spatial features and the time series features, constructing a spatio-temporal adaptability coefficient, then concatenating the context vector composed of the crop growth stage one-hot encoding and the crop physiological load factor through a gating network to generate a dynamic weight, and performing adaptive weight correction according to the spatio-temporal adaptability coefficient, finally weighting and fusing the spatial features and the time series features to obtain spatio-temporal fusion features; In step S24, the multi-target prediction output, specifically, a multi-task learning model is used to perform multi-output regression prediction on the spatio-temporal fusion features, and an agricultural proportion constraint and a spatio-temporal consistency constraint are introduced to construct a comprehensive loss function to correct the prediction results, and obtain the water and fertilizer demand prediction information; The agricultural proportion constraint is used to ensure that the predicted values of nitrogen, phosphorus and potassium satisfy the agricultural empirical proportion relationship; the spatio-temporal consistency constraint is used to maintain the consistency of the spatial features and the time series features in the spatio-temporal fusion features; Step S3: planting optimization layer construction, the planting optimization layer construction is used to generate an optimal water and fertilizer management strategy considering yield, resource and environmental benefits, specifically, according to the water and fertilizer prediction information and the spatio-temporal fusion data set, a planting optimization layer construction method combining digital twin and multi-objective optimization is adopted to obtain the optimal water and fertilizer management strategy, comprising the following steps: step S31 decision space initialization, step S32 digital twin deduction, step S33 multi-objective optimization solution and step S34 preference interactive decision; In step S31, the decision space is initialized, specifically by constructing a water and fertilizer decision vector, setting a dynamic search interval, and introducing a weather scenario set to construct a multi-objective optimization function, setting multiple constraint conditions, and obtaining a multi-objective decision space model; In step S32, the digital twin is deduced, which is used to deduce the crop growth process according to the candidate water and fertilizer management strategy, specifically by constructing a digital twin including a crop growth mechanism model, a data agent model, and a physical consistency correction module, receiving field environment monitoring data, future weather forecast, and candidate water and fertilizer management strategy as input, performing time series deduction on yield and nutrient residue, and performing constraint correction through the physical consistency correction module to obtain crop growth simulation results; In step S33, the multi-objective optimization is solved, which is used to obtain a stable and executable water and fertilizer management strategy in the presence of multiple constraint conditions and weather forecast errors, specifically by combining the decision space initialization and the digital twin deduction, introducing a dynamic penalty mechanism, a robust fitness mechanism, and a constraint-priority double selection mechanism based on a multi-objective evolutionary algorithm, and adaptively optimizing the candidate water and fertilizer management strategy to obtain a candidate water and fertilizer management strategy set; In step S34, the preference interactive decision is used to generate a final water and fertilizer management strategy from the strategy set according to user preferences, specifically by obtaining user preference weights for yield, water saving, and environmental risk, weighting and reordering the candidate water and fertilizer management strategy set, and selecting the optimal water and fertilizer management strategy that meets the preference constraints; Step S4: agricultural planting management. 2.The intelligent agricultural planting management method according to claim 1, characterized in that: In step S31, the water and fertilizer decision vector includes irrigation amount and nitrogen, phosphorus, and potassium application amount; the dynamic search interval is set to construct a feasible search range centered on the predicted value and adaptively adjusted with the growth stage; The multi-objective optimization function is used to evaluate yield improvement, water resource utilization efficiency, and nitrogen leaching risk simultaneously in the presence of weather forecast errors, including minimizing the negative yield objective function, minimizing the water consumption objective function, and minimizing the nitrogen leaching risk objective function; The multiple constraint conditions are set, specifically for soil moisture content, nutrient allocation, and equipment capacity constraints. 3.The intelligent agricultural planting management method according to claim 2, characterized in that: In step S33, the dynamic penalty mechanism uses an exponential penalty factor that increases with the number of iterations to gradually suppress constraint violation behavior; the robust fitness mechanism constructs a robust fitness by introducing a weather scenario set to evaluate the robustness of the candidate water and fertilizer management strategy under multiple weather scenarios; the constraint-priority double selection mechanism specifically selects water and fertilizer management strategies that fully meet the multiple constraint conditions in the non-dominated sorting process, and then adaptively sorts and selects candidate solutions according to the comprehensive fitness to realize the iterative evolution of the candidate water and fertilizer management strategy set.

4. The intelligent agricultural planting management method according to claim 3, characterized in that: In step S32, the crop growth mechanism model is used to simulate water transport, evaporation process, soil moisture content change, and potential growth state of crops; The data agent model is a multi-output regression model based on machine learning, which is used to preliminarily predict yield and soil nutrient residue.

5. The intelligent agricultural planting management method according to claim 4, characterized in that: In step S4, the agricultural planting management, specifically, by executing steps S1 to S3, the data layer, the water and fertilizer prediction layer and the planting optimization layer are integrated to construct a water and fertilizer optimization model for agricultural planting management, water and fertilizer optimization is performed, optimal water and fertilizer management strategies are generated, field-level control instructions are issued to the water and fertilizer integrated control system, and execution effect monitoring is performed to generate planting management feedback reports. In step S1, the data layer is constructed, specifically, by collecting multi-channel remote sensing image data, field environment monitoring data, future weather forecasts and crop growth information, and performing preprocessing, spatio-temporal alignment and standardization processing to obtain a spatio-temporal fusion data set.

6. An intelligent agricultural planting management system for implementing the intelligent agricultural planting management method according to any one of claims 1-5, characterized in that: It comprises a data establishment module, a water and fertilizer prediction module, a planting optimization module and an agricultural planting management module.

7. The intelligent agricultural planting management system according to claim 6, characterized in that: The data establishment module is used for data layer construction, and through data layer construction, a spatio-temporal fusion data set is obtained, and the spatio-temporal fusion data set is sent to the water and fertilizer prediction module and the planting optimization module; The water and fertilizer prediction module is used for water and fertilizer prediction layer construction, and through water and fertilizer prediction layer construction, water and fertilizer demand prediction information is obtained, and the water and fertilizer demand prediction information is sent to the planting optimization module; The planting optimization module is used for planting optimization layer construction, and through planting optimization layer construction, optimal water and fertilizer management strategies are obtained, and the optimal water and fertilizer management strategies are sent to the agricultural planting management module; The agricultural planting management module is used for agricultural planting management to obtain field-level control instructions and planting management feedback reports.

Images