Crop water and fertilizer decision determination method and device, electronic equipment and storage medium

Abstract

The application provides a crop water and fertilizer decision determination method and device, electronic equipment and a storage medium, and belongs to the technical field of agricultural data processing.The method comprises the following steps: obtaining crop model fitting parameters of a crop based on a water and fertilizer decision model; wherein the water and fertilizer decision model is obtained based on sample basic data and sample labels of sample crops and by training on the basis of a preset fusion model; obtaining a nitrogen regulation amount and an irrigation amount of the crop based on the crop model fitting parameters and preset parameters; and determining the water and fertilizer decision of the crop based on the nitrogen regulation amount and the irrigation amount.The application obtains the crop model fitting parameters of the crop through the water and fertilizer decision model, considers more influencing factors, improves the efficiency and accuracy of determining the crop model fitting parameters, and further improves the accuracy of determining the nitrogen regulation amount and the irrigation amount of the crop, which is beneficial to improving the accuracy of the water and fertilizer decision of the crop.

Description

Technical Field

[0001] This invention relates to the field of agricultural data processing technology, and in particular to a method, apparatus, electronic device, and storage medium for determining water and fertilizer decisions for crops. Background Technology

[0002] Because crop growth is influenced by multiple factors such as light, temperature, water, air, and heat, its growth and development is the result of the combined effects of all environmental conditions and cultivation measures. Even when managed according to the cultivation management plan designed before sowing, its growth often deviates from the optimal state. Therefore, it is necessary to monitor and analyze the crop growth status during the production process, and on this basis, to make real-time adjustments to water and fertilizer management. Crop water and fertilizer decisions are a crucial aspect of crop cultivation management.

[0003] Existing methods for determining water and fertilizer application in crops require multi-year, multi-location field trials with different water and fertilizer gradients. These trials involve curve fitting and parameter optimization of the experimental data to determine crop production. However, these methods suffer from limited consideration of factors and poor scalability, leading to inaccurate water and fertilizer decisions. Summary of the Invention

[0004] This invention provides a method, apparatus, electronic device, and storage medium for determining water and fertilizer decisions for crops, in order to solve the defects of inaccurate water and fertilizer decisions for crops in the prior art and improve the accuracy of water and fertilizer decisions.

[0005] In a first aspect, the present invention provides a method for determining water and fertilizer decisions for crops, comprising: obtaining crop model fitting parameters for crops based on a water and fertilizer decision model; wherein the water and fertilizer decision model is trained on a preset fusion model based on sample basic data and sample labels of sample crops, the sample basic data including historical growth environment data and historical growth cycle data, and the sample labels including sample yield, sample nitrogen regulation amount, or sample water consumption of sample crops; obtaining the nitrogen regulation amount and irrigation amount of crops based on the crop model fitting parameters and preset parameters; and determining the water and fertilizer decisions for crops based on the nitrogen regulation amount and irrigation amount.

[0006] According to the water and fertilizer decision-making method for crops provided by the present invention, the water and fertilizer decision-making model is obtained based on the following steps: obtaining historical growth environment data based on meteorological data and soil baseline data of sample crops during their growth process; obtaining historical growth cycle data based on crop canopy images, 3D point clouds, RGB images, spectral image data, thermal infrared data, and hyperspectral data during the crop growth process to obtain sample baseline data; obtaining labeled sample baseline data based on sample labels and sample baseline data; training and testing a preset fusion model based on the sample baseline data; and determining that the preset fusion model training is complete when the error of the prediction result of the preset fusion model is less than a set threshold. An initial water and fertilizer decision model was obtained. The initial water and fertilizer decision model was then subjected to dimensionality reduction to varying degrees to simplify the input and output data, resulting in a new water and fertilizer decision model. When the input data is the crop yield estimation coefficient, the corresponding output data is the target crop yield. When the input data is the crop nitrogen nutrient index, the corresponding output data is the relative crop yield. When the input data is the crop's suitable nitrogen accumulation, the corresponding output data is the crop's nitrogen supplementation regulation amount. When the input data is the nitrogen nutrient index, the corresponding output data is the nitrogen supplementation regulation amount. When the input data is the reference crop's water evapotranspiration, the corresponding output data is the crop's actual water consumption.

[0007] The method for determining water and fertilizer decisions for crops according to the present invention, based on a water and fertilizer decision model, obtains crop model fitting parameters for crops, including: measuring the growth of crops to obtain multiple input data; inputting each input data into the water and fertilizer decision model to obtain the output data of the water and fertilizer decision model; constructing a data set based on each input data and the corresponding output data; constructing transformation information of the input data and the corresponding output data based on the initial crop model fitting parameters; and inputting multiple data sets into the transformation information to update the initial crop model fitting parameters to obtain crop model fitting parameters.

[0008] According to the method for determining water and fertilizer application for crops provided by the present invention, the preset parameters include the crop's nitrogen uptake per 100 kg of grain, the soil's basic nitrogen supply, the basic nitrogen application rate, and the nitrogen fertilizer utilization rate. The method obtains the nitrogen topdressing regulation amount for crops based on crop model fitting parameters and preset parameters, including: updating the third transformation information based on the third crop model fitting parameters, inputting the yield estimation coefficient into the updated third transformation information to update the target yield; or updating the fourth transformation information based on the fourth crop model fitting parameters, inputting the nitrogen nutrient index into the updated fourth transformation information to update the relative yield, and updating the target yield based on the extreme value of the relative yield and the maximum yield of the crop; the third crop model fitting parameters are the crop model fitting parameters of the yield estimation coefficient and the target yield, the fourth crop model fitting parameters are the crop model fitting parameters of the nitrogen nutrient index and the target yield, the third transformation information is the transformation information corresponding to the third crop model fitting parameters, and the fourth transformation information is the transformation information corresponding to the fourth crop model fitting parameters; determining the crop's nitrogen requirement based on the updated target yield and the crop's nitrogen uptake per 100 kg of grain; and determining the nitrogen topdressing regulation amount based on the nitrogen requirement, the soil's basic nitrogen supply, the basic nitrogen application rate, and the nitrogen fertilizer utilization rate.

[0009] According to the water and fertilizer decision-making method for crops provided by the present invention, the preset parameters include the real-time soil volumetric water content, irrigation depth, soil moisture content, and the percentage of real-time soil volumetric water content. The method obtains the irrigation amount based on crop model fitting parameters and preset parameters, including: updating the fifth transformation information based on the fifth crop model fitting parameters; inputting water evapotranspiration into the updated fifth transformation information to update the actual water consumption; the fifth crop model fitting parameters are the crop model fitting parameters for water evapotranspiration and actual water consumption; and the fifth transformation information is the transformation information corresponding to the fifth crop model fitting parameters; and determining the irrigation amount based on the updated actual water consumption, real-time soil volumetric water content, irrigation depth, soil moisture content, and the percentage of real-time soil volumetric water content.

[0010] According to the method for determining water and fertilizer decisions for crops provided by the present invention, determining the amount of nitrogen topdressing further includes: updating the nitrogen nutrient index based on the fitting parameters of a first crop model, wherein the fitting parameters of the first crop model are crop model fitting parameters of the appropriate nitrogen accumulation and the amount of nitrogen topdressing; and updating the amount of nitrogen topdressing based on the fitting parameters of a second crop model and the updated nitrogen nutrient index, wherein the fitting parameters of the second crop model are crop model fitting parameters of the nitrogen nutrient index and the amount of nitrogen topdressing.

[0011] According to the method for determining water and fertilizer decisions for crops provided by the present invention, the nitrogen nutrition index is determined by: determining the critical nitrogen concentration of the crop based on the aboveground biomass of the crop and the fitting parameters of the initial first crop model; and obtaining the nitrogen nutrition index based on the ratio of the actual nitrogen concentration of the crop to the critical nitrogen concentration.

[0012] Secondly, the present invention also provides a device for determining water and fertilizer decisions for crops, comprising: a crop model fitting parameter determination module, used to obtain crop model fitting parameters for crops based on a water and fertilizer decision model; wherein, the water and fertilizer decision model is trained based on a preset fusion model and on sample basic data and sample labels of sample crops, the sample basic data including historical growth environment data and historical growth cycle data, and the sample labels including sample yield, sample nitrogen regulation amount, or sample water consumption of sample crops; a nitrogen regulation amount and irrigation amount determination module, used to obtain nitrogen regulation amount and irrigation amount of crops based on crop model fitting parameters and preset parameters; and a water and fertilizer decision determination module, used to determine water and fertilizer decisions for crops based on nitrogen regulation amount and irrigation amount.

[0013] Thirdly, the present invention 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 program to implement the water and fertilizer decision determination method for any of the above-described crops.

[0014] Fourthly, 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 water and fertilizer decision determination method for any of the above-described crops.

[0015] This invention provides a method, apparatus, electronic device, and storage medium for determining water and fertilizer decisions for crops. Based on a water and fertilizer decision-making model, it obtains crop model fitting parameters. The water and fertilizer decision-making model is trained on a pre-defined fusion model using sample crop data and labels. The sample data includes historical growth environment data and historical growth cycle data. The sample labels include sample crop yield, sample nitrogen regulation amount, or sample water consumption. Based on the crop model fitting parameters and pre-defined parameters, the nitrogen regulation amount and irrigation amount for the crop are obtained. Based on the nitrogen regulation amount and irrigation amount, the water and fertilizer decision for the crop is determined. This invention, through the water and fertilizer decision-making model, obtains crop model fitting parameters, considers more influencing factors, improves the efficiency and accuracy of determining crop model fitting parameters, and thus improves the accuracy of determining the nitrogen regulation amount and irrigation amount for the crop, thereby contributing to the improvement of the accuracy of water and fertilizer decisions for crops. Attached Figure Description

[0016] 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.

[0017] Figure 1 This is a flowchart illustrating the water and fertilizer decision-making method for crops provided by the present invention.

[0018] Figure 2 This is a schematic diagram of the water and fertilizer decision-making device for crops provided by the present invention;

[0019] Figure 3 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0021] The following is combined Figures 1-3 This invention describes the method, apparatus, and electronic equipment for determining water and fertilizer decisions for crops provided in embodiments of the present invention.

[0022] Figure 1 This is a flowchart illustrating the water and fertilizer decision-making method for crops provided by the present invention, as shown below. Figure 1 As shown, the method for determining water and fertilizer decisions for crops includes steps S100 to S300, each step of which is detailed below:

[0023] S100: Based on the water and fertilizer decision model, obtain crop model fitting parameters for crops.

[0024] Among them, the water and fertilizer decision model is trained based on the sample basic data and sample labels of the sample crops, on the basis of the pre-set fusion model. The sample basic data includes historical growth environment data and historical growth cycle data, and the sample labels include the sample yield of the sample crops, the sample nitrogen regulation amount, or the sample water consumption.

[0025] The water and fertilizer decision model is obtained based on the following steps: historical growth environment data are obtained based on meteorological data and soil basic data of the sample crops during their growth process; historical growth cycle data are obtained based on 3D point cloud, RGB image, spectral image data, thermal infrared data, and hyperspectral data of the crop canopy during the crop growth process, to obtain sample basic data; tagged sample basic data are obtained based on sample labels and sample basic data; the preset fusion model is trained and tested based on the sample basic data; when the error of the prediction result of the preset fusion model is less than a set threshold, the preset fusion model is determined to be trained successfully, and the initial water and fertilizer decision model is obtained. The initial water and fertilizer decision model was dimensionality reduced to different degrees to simplify the input and output data, resulting in the final water and fertilizer decision model. When the input data is the crop yield estimation coefficient, the corresponding output data is the target crop yield. When the input data is the crop nitrogen nutrient index, the corresponding output data is the relative crop yield. When the input data is the crop's suitable nitrogen accumulation, the corresponding output data is the crop's nitrogen supplementation regulation amount. When the input data is the nitrogen nutrient index, the corresponding output data is the nitrogen supplementation regulation amount. When the input data is the reference crop's water evapotranspiration, the corresponding output data is the crop's actual water consumption.

[0026] The pre-defined fusion model is a fusion model built using multi-source data fusion and artificial intelligence algorithms such as Transformer. Using Transformer as a feature extractor, features are extracted from RGB images, 3D point clouds, spectral image data, thermal infrared data, and hyperspectral data. The extracted features are then cross-fused to construct a deep learning-based water and fertilizer decision-making model.

[0027] Meteorological and soil baseline data (including soil moisture data) for the entire growth period of the sample crops were acquired to obtain historical growth environment data. Historical growth cycle data was obtained by acquiring crop canopy images, 3D point clouds, RGB images, spectral image data, thermal infrared data, and hyperspectral data during the crop growth process using image acquisition devices (e.g., drones). Based on the historical growth environment and historical growth cycle data of the sample crops, the sample baseline data of the sample crops were determined. Plant height, leaf area index, and other data of the sample crops were measured at key growth stages to calculate the sample yield. The sample nitrogen application rate and sample water consumption during the growth process were calculated.

[0028] When the sample label is the sample yield, the basic sample data is labeled based on the sample yield to obtain the labeled basic sample data. The pre-set fusion model is trained and tested based on the labeled basic sample data to obtain the initial water and fertilizer decision model. The initial water and fertilizer decision model is then dimensionality-reduced to obtain the final water and fertilizer decision model. At this point, when the input data of the water and fertilizer decision model is the crop yield estimation coefficient, the output data is the target yield. When the input data of the water and fertilizer decision model is the nitrogen nutrient index, the output data is the relative yield.

[0029] When the sample label is the sample water consumption, the basic sample data is labeled according to the sample water consumption to obtain labeled basic sample data. Based on the labeled basic sample data, the preset fusion model is trained, tested, and its dimensionality reduced to obtain the water and fertilizer decision model. At this point, when the input data of the water and fertilizer decision model is the crop's water evapotranspiration, the output data is the actual water consumption.

[0030] When the sample label is the sample nitrogen supplementation regulation amount, the sample basic data is labeled according to the sample nitrogen supplementation regulation amount to obtain the labeled sample technical data. Based on the labeled sample basic data, the preset fusion model is trained, tested, and its dimensionality reduced to obtain the water and fertilizer decision model. At this point, when the input data of the water and fertilizer decision model is the nitrogen nutrient index or suitable nitrogen accumulation amount of the crop, the output data is the nitrogen supplementation regulation amount.

[0031] Based on the water and fertilizer decision-making model, crop model fitting parameters are obtained. Specifically, the growth of crops is measured to obtain multiple input data. Each input data is input into the water and fertilizer decision-making model to obtain the output data of the water and fertilizer decision-making model. Data sets are constructed based on each input data and the corresponding output data. Based on the initial crop model fitting parameters, transformation information of input data and corresponding output data is constructed. Multiple data sets are input into the transformation information to update the initial crop model fitting parameters, thus obtaining the crop model fitting parameters.

[0032] Nitrogen topdressing refers to the application of additional nitrogen fertilizer to the soil to meet the nitrogen requirements of crops during their growth process, based on factors such as the crop's growth stage and soil nitrogen content. Nitrogen is one of the key nutrients essential for plant growth, and nitrogen topdressing has a significant impact on plant growth, development, yield, and quality.

[0033] The crop model fitting parameters include the fitting parameters for the first, second, third, fourth, and fifth crop models. The transformation information includes the first, second, third, fourth, and fifth transformation information.

[0034] The first crop model fitting parameters are used to characterize the first conversion information between suitable nitrogen accumulation and topdressing regulation. The second crop model fitting parameters are used to characterize the second conversion information between the nitrogen nutrient index and topdressing regulation. The third crop model fitting parameters are used to characterize the third conversion information between the target yield and the yield estimation coefficient. The fourth crop model fitting parameters are used to characterize the fourth conversion information between the nitrogen nutrient index and the target yield. The fifth crop model fitting parameters are used to characterize the fifth conversion information between water evapotranspiration and actual water consumption.

[0035] When the input data is the crop yield estimation coefficient, the corresponding output data is the target yield of the crop. The yield estimation coefficient is used to predict the target yield of the crop. The formulas for expressing the target yield and the yield estimation coefficient are as follows:

[0036]

[0037] Where INSEY is the yield estimation coefficient, NDVI is the normalized difference in vegetation index of the crop as measured in actual measurements, GADD is the cumulative growth degree days of the crop from the sowing date, PGY is the target yield, and a3 and b3 are the fitting parameters of the third crop model. This is the third type of conversion information.

[0038] Each crop yield estimation coefficient (INSEY) is input into the water and fertilizer decision model to obtain the target yield (PGY) output by the model. Multiple sets of yield estimation coefficients and target yields are then input into the third transformation information to construct the objective equation. Solving the objective equation yields the fitting parameters for the third crop model.

[0039] When the input data is the nitrogen nutrient index (NNI) of crops, the corresponding output data is the relative yield. The NNI is used to predict relative yield. The amount of nitrogen applied during the crop's growth period is related to the NNI. If NNI < 1, it indicates that the crop is nitrogen-deficient and the amount of nitrogen applied needs to be increased. If NNI > 1, it indicates that the crop is nitrogen-excessive and the amount of nitrogen applied needs to be reduced. If NNI = 1, it indicates that nitrogen nutrition is adequate and no adjustment of the amount of nitrogen applied is needed.

[0040] The formulas for expressing the nitrogen nutrient index and relative yield are as follows:

[0041]

[0042] Where RY is relative yield, NNI is nitrogen nutrient index, a4 and b4 are the fitting parameters of the fourth crop model, YPSI is the maximum yield of crops in nitrogen-sufficient areas, PGY is the target yield, and NDVI is the normalized vegetation index of the actual measured crops. opt The normalized vegetation index (NNI) for suitable crops is RY = a4 × NNI + b4, which represents the fourth transformation information.

[0043] Each nitrogen nutrient index (NNI) of the crop is input into the water and fertilizer decision model to obtain the relative yield (RY). Multiple sets of nitrogen nutrient indices and relative yields are input into the fourth transformation information to construct the objective equation. The objective equation is then solved to obtain the fitting parameters for the fourth crop model.

[0044] When the input data is the optimal nitrogen accumulation level, the corresponding output data is the nitrogen supplementation regulation amount. When the input data is the nitrogen nutrient index, the corresponding output data is the nitrogen supplementation regulation amount for crops.

[0045] Determine the nitrogen nutrient index: Based on the aboveground biomass of the crop and the fitting parameters of the initial first crop model, determine the critical nitrogen concentration of the crop; based on the ratio of the actual nitrogen concentration of the crop to the critical nitrogen concentration, obtain the nitrogen nutrient index.

[0046] The formulas for calculating the nitrogen nutrient index and the amount of nitrogen supplementation are as follows:

[0047]

[0048] Where a1 and b1 are the fitting parameters of the first crop model, N cnc The critical nitrogen concentration, N anc The actual nitrogen concentration is the measured value, DM is the aboveground biomass, NNI is the nitrogen nutrient index, and N is the nitrogen content index. and To track nitrogen regulation, N na The actual nitrogen accumulation is expressed in kilograms per hectare (kg / ha), where m and n are the fitting parameters for the second crop model, and N... cna The optimal nitrogen accumulation level for crops is defined as NUE, which represents nitrogen fertilizer utilization efficiency (empirical value). For the first transformation information, N and =m×NNI+n is the second conversion information.

[0049] Each suitable nitrogen accumulation level for the crop is input into the water and fertilizer decision model to obtain the nitrogen topdressing regulation amount. Multiple sets of suitable nitrogen accumulation levels and nitrogen topdressing regulation amounts are input into the first transformation information to construct the objective equation. The objective equation is solved to obtain the fitting parameters of the first crop model.

[0050] Each nitrogen nutrient index of the crop is input into the water and fertilizer decision model to obtain the nitrogen topdressing regulation amount. Multiple sets of nitrogen nutrient indices and nitrogen topdressing regulation amounts are input into the second transformation information to construct the objective equation. The objective equation is solved to obtain the fitting parameters of the second crop model.

[0051] When the input data is water evapotranspiration, the corresponding output data is the actual water consumption. The formula for calculating the actual water consumption is:

[0052] ET=K c ×ET0;

[0053] Where ET is the actual water consumption, and Kc Here are the fitting parameters for the fifth crop model, where ET0 is the water evapotranspiration of the reference crop calculated based on the Penman formula, and ET = K. c ×ET0 is the fifth conversion information.

[0054] The evapotranspiration of each crop is input into the water and fertilizer decision model to obtain the actual water consumption. Multiple sets of evapotranspiration and actual water consumption are input into the fifth transformation information to construct the objective equation. The objective equation is then solved to obtain the fitting parameters of the fifth crop model.

[0055] S200: Obtains the amount of nitrogen supplementation and irrigation for crops based on crop model fitting parameters and preset parameters.

[0056] The preset parameters include nitrogen uptake per 100 kg of crop grain, basic soil nitrogen supply, basic nitrogen application, and nitrogen fertilizer utilization rate. The nitrogen topdressing regulation amount for crops is obtained based on crop model fitting parameters and preset parameters. Specifically, the third transformation information is updated based on the third crop model fitting parameters, and the yield estimation coefficient is input into the updated third transformation information to update the target yield; or the fourth transformation information is updated based on the fourth crop model fitting parameters, and the nitrogen nutrient index is input into the updated fourth transformation information to update the relative yield. The target yield is updated based on the extreme value of the relative yield and the maximum yield of the crop. The third crop model fitting parameters are the crop model fitting parameters for the yield estimation coefficient and the target yield; the fourth crop model fitting parameters are the crop model fitting parameters for the nitrogen nutrient index and the target yield; the third transformation information is the transformation information corresponding to the third crop model fitting parameters; and the fourth transformation information is the transformation information corresponding to the fourth crop model fitting parameters. Based on the updated target yield and nitrogen uptake per 100 kg of crop grain, the nitrogen requirement of the crop is determined. Based on the nitrogen requirement, basic soil nitrogen supply, basic nitrogen application, and nitrogen fertilizer utilization rate, the nitrogen topdressing regulation amount is determined.

[0057] The third transformation information is updated based on the fitting parameters of the third crop model. The yield estimation coefficient is input into the updated third transformation information to obtain the updated target yield. The fourth transformation information is updated based on the fitting parameters of the fourth crop model. The yield estimation coefficient is input into the updated fourth transformation information to obtain the updated relative yield. The updated target yield is obtained by calculating the product of the updated relative yield and the maximum yield.

[0058] The formula for calculating the nitrogen supplementation amount is:

[0059]

[0060] Where, N and To determine the amount of nitrogen to be adjusted, NUE is nitrogen fertilizer utilization rate, PGY is target yield, GNU is nitrogen requirement, and N... abs Nitrogen uptake per 100 kilograms of crop grains, expressed in kg (N).base Basal nitrogen application rate, in kg / ha, N soi The basic nitrogen supply to the soil is expressed in kg / ha.

[0061] Determining the nitrogen topdressing regulation amount also includes: updating the nitrogen nutrient index based on the fitting parameters of the first crop model, where the fitting parameters of the first crop model are the crop model fitting parameters of the appropriate nitrogen accumulation and the nitrogen topdressing regulation amount; and updating the nitrogen topdressing regulation amount based on the fitting parameters of the second crop model and the updated nitrogen nutrient index, where the fitting parameters of the second crop model are the crop model fitting parameters of the nitrogen nutrient index and the nitrogen topdressing regulation amount.

[0062] When the input data is the nitrogen nutrient index and the output data is the nitrogen topdressing regulation amount, the nitrogen nutrient index is updated based on the fitting parameters of the first crop model. The nitrogen topdressing regulation amount is then updated based on the updated nitrogen nutrient index and the fitting parameters of the second crop model.

[0063] The preset parameters include the real-time soil volumetric water content of the crop, irrigation depth, soil moisture content, and the percentage of real-time soil volumetric water content. Irrigation amount is obtained based on the crop model fitting parameters and preset parameters. Specifically, the fifth transformation information is updated based on the fifth crop model fitting parameters. Water evapotranspiration is input into the updated fifth transformation information to update the actual water consumption. The fifth crop model fitting parameters are the crop model fitting parameters for water evapotranspiration and actual water consumption, and the fifth transformation information is the transformation information corresponding to the fifth crop model fitting parameters. The irrigation amount is determined based on the updated actual water consumption, real-time soil volumetric water content, irrigation depth, soil moisture content, and the percentage of real-time soil volumetric water content.

[0064] The fifth transformation information is updated based on the fitting parameters of the fifth crop model. Evapotranspiration is input into the updated fifth transformation information to obtain the updated actual water consumption. Based on the updated actual water consumption, real-time soil volumetric water content, irrigation depth, soil moisture content, and the percentage of real-time soil volumetric water content, the irrigation amount is determined. The formula for calculating the irrigation amount is:

[0065] Ir=ET-(SW-p×SW FC )×Depth×100;

[0066] Where Ir is the irrigation amount, ET is the actual water consumption, SW is the soil moisture content, and SW is the soil moisture content. FC ρ represents the real-time soil volumetric water content, p represents the percentage of real-time soil volumetric water content, and Depth represents the irrigation depth.

[0067] S300: Determines crop water and fertilizer decisions based on nitrogen supplementation and irrigation amounts.

[0068] The amount of fertilizer to be applied to crops is determined based on the amount of nitrogen applied. The amount of water required for crops is determined based on the amount of irrigation. Water and fertilizer decisions for crops are then made based on the amount of fertilizer and water applied.

[0069] The water and fertilizer decision-making method for crops provided in this invention is based on a water and fertilizer decision-making model to obtain crop model fitting parameters. The water and fertilizer decision-making model is trained on a pre-defined fusion model using sample crop basic data and sample labels. The sample basic data includes historical growth environment data and historical growth cycle data, and the sample labels include sample crop yield, sample nitrogen regulation amount, or sample water consumption. The nitrogen regulation amount and irrigation amount for the crop are obtained based on the crop model fitting parameters and the pre-defined parameters. The water and fertilizer decision for the crop is then determined based on the nitrogen regulation amount and irrigation amount. This invention, through the water and fertilizer decision-making model, obtains crop model fitting parameters, considers more influencing factors, improves the efficiency and accuracy of determining crop model fitting parameters, and thus improves the accuracy of determining the nitrogen regulation amount and irrigation amount for the crop, thereby improving the accuracy of water and fertilizer decision-making for crops.

[0070] To further explain the crop water and fertilizer decision-making method provided in the embodiments of the present invention, the following embodiments are used for illustration:

[0071] (1) Construct a water and fertilizer decision-making model. The input data of the water and fertilizer decision-making model is the yield estimation coefficient, and the output data is the target yield. Input multiple yield estimation coefficients into the water and fertilizer decision-making model in sequence to obtain multiple target yields. Construct a third data set based on the yield estimation coefficients and target yields. Construct third transformation information of yield estimation coefficients and target yields based on the initial third crop model fitting parameters. Input multiple third data sets into the third transformation information and solve the initial third crop model fitting parameters to obtain the third crop model fitting parameters. Update the target yield based on the third crop model fitting parameters. Based on the updated target yield and the nitrogen uptake per 100 kg of crop grain, determine the nitrogen requirement of the crop. Calculate the topdressing nitrogen regulation amount based on the nitrogen requirement, soil basic nitrogen supply, basic nitrogen application, and nitrogen fertilizer utilization rate.

[0072] Alternatively, a water and fertilizer decision-making model can be constructed, with nitrogen nutrient index as input and target yield as output. Multiple nitrogen nutrient indices are sequentially input into the water and fertilizer decision-making model to obtain multiple target yields. A fourth data set is constructed based on the nitrogen nutrient index and target yield. A fourth transformation information set between the nitrogen nutrient index and target yield is constructed based on the initial fourth fitting parameters. Multiple fourth data sets are input into the fourth transformation information to solve for the initial fourth crop model fitting parameters, obtaining the fourth crop model fitting parameters. The target yield is updated based on the fourth crop model fitting parameters. Based on the updated target yield and the nitrogen uptake per 100 kg of crop grain, the crop's nitrogen requirement is determined. The amount of topdressing nitrogen is calculated based on the nitrogen requirement, soil basal nitrogen supply, basal nitrogen application, and nitrogen fertilizer utilization rate.

[0073] Alternatively, a water and fertilizer decision-making model can be constructed. The input data for this model are the optimal nitrogen accumulation rate or nitrogen nutrient index, and the output data is the nitrogen topdressing regulation rate. Multiple optimal nitrogen accumulation rates are sequentially input into the water and fertilizer decision-making model to obtain multiple nitrogen topdressing regulation rates. A first data set is constructed based on the optimal nitrogen accumulation rate and the nitrogen topdressing regulation rate. Based on the initial first fitting parameters, first transformation information for the optimal nitrogen accumulation rate and the nitrogen topdressing regulation rate is constructed. Multiple first data sets are input into the first transformation information, and the initial first crop model fitting parameters are solved to obtain the first crop model fitting parameters. Multiple nitrogen nutrient indices are sequentially input into the water and fertilizer decision-making model to obtain multiple nitrogen topdressing regulation rates. A second data set is constructed based on the nitrogen nutrient index and the nitrogen topdressing regulation rate. Based on the initial second fitting parameters, second transformation information for the nitrogen nutrient index and the nitrogen topdressing regulation rate is constructed. Multiple second data sets are input into the second transformation information, and the initial second crop model fitting parameters are solved to obtain the second crop model fitting parameters. The nitrogen nutrient index is updated based on the first crop model fitting parameters, and the nitrogen topdressing regulation rate is updated based on the updated nitrogen nutrient index and the second crop model fitting parameters.

[0074] (2) Construct a water and fertilizer decision-making model. The input data of the water and fertilizer decision-making model is water evapotranspiration, and the output data is actual water consumption. Input multiple water evapotranspiration values ​​into the water and fertilizer decision-making model sequentially to obtain multiple actual water consumption values. Construct a fifth data set based on water evapotranspiration and actual water consumption. Based on the initial fifth fitting parameters, construct the fifth transformation information of water evapotranspiration and actual water consumption. Input multiple fifth data sets into the fifth transformation information and solve the initial fifth crop model fitting parameters to obtain the fifth crop model fitting parameters. Update the actual water consumption based on the fifth crop model fitting parameters. Determine the irrigation amount based on the updated actual water consumption, real-time soil volumetric water content, irrigation depth, soil moisture content, and the percentage of real-time soil volumetric water content.

[0075] (3) Determine the water and fertilizer decisions for crops based on irrigation volume and nitrogen regulation.

[0076] Traditional water and fertilizer decision-making models are direct input-output systems with a black box in between. These models suffer from poor interpretability and weak mechanistic understanding, making them difficult to integrate with parameters such as fertilization and irrigation in production management for effective control. The water and fertilizer decision-making method of this invention is interpretable throughout its entire process. By determining crop model fitting parameters through a water and fertilizer decision-making model, more data and factors are considered, resulting in more accurate crop model fitting parameters. This invention, while maintaining the mechanistic understanding and interpretability of the water and fertilizer decision-making model, incorporates more factors into the decision-making process through AI algorithms (water and fertilizer decision-making model), automatically determining crop model fitting parameters and significantly improving the accuracy of water and fertilizer decisions.

[0077] This invention also provides a device for determining water and fertilizer usage for crops, such as... Figure 2 As shown, Figure 2 This is a schematic diagram of the crop water and fertilizer decision-making device provided by the present invention. It should be noted that the crop water and fertilizer decision-making device provided in this embodiment can execute the crop water and fertilizer decision-making method described in any of the above embodiments during operation; however, this embodiment will not elaborate further on this.

[0078] Reference Figure 2 The present invention provides a water and fertilizer decision-making device for crops, comprising:

[0079] The crop model fitting parameter determination module 201 is used to obtain crop model fitting parameters for crops based on the water and fertilizer decision model. The water and fertilizer decision model is trained on the basis of the sample basic data and sample labels of the sample crops, based on the preset fusion model. The sample basic data includes historical growth environment data and historical growth cycle data, and the sample labels include the sample yield, sample nitrogen regulation amount, or sample water consumption of the sample crops.

[0080] The nitrogen application rate and irrigation amount determination module 202 is used to obtain the nitrogen application rate and irrigation amount of crops based on crop model fitting parameters and preset parameters.

[0081] The water and fertilizer decision determination module 203 is used to determine the water and fertilizer decisions for crops based on the amount of nitrogen supplementation and irrigation.

[0082] The water and fertilizer decision-making device for crops provided in this invention obtains crop model fitting parameters based on a water and fertilizer decision-making model. The water and fertilizer decision-making model is trained on a preset fusion model using sample crop basic data and sample labels. The sample basic data includes historical growth environment data and historical growth cycle data, and the sample labels include sample crop yield, sample nitrogen regulation amount, or sample water consumption. The nitrogen regulation amount and irrigation amount for the crop are obtained based on the crop model fitting parameters and preset parameters. The water and fertilizer decision for the crop is then determined based on the nitrogen regulation amount and irrigation amount. This invention, through the water and fertilizer decision-making model, obtains crop model fitting parameters, considers more influencing factors, improves the efficiency and accuracy of determining crop model fitting parameters, and thus improves the accuracy of determining the nitrogen regulation amount and irrigation amount for the crop, thereby improving the accuracy of water and fertilizer decision-making for crops.

[0083] In one embodiment, the crop model fitting parameter determination module 201 is used to: obtain historical growth environment data based on meteorological data and soil baseline data of the sample crop during its growth process; obtain historical growth cycle data based on crop canopy images, 3D point clouds, RGB images, spectral image data, thermal infrared data, and hyperspectral data during the crop's growth process to obtain sample baseline data; obtain labeled sample baseline data based on sample labels and sample baseline data; train and test a preset fusion model based on the sample baseline data; and determine that the preset fusion model training is complete when the error of the prediction result of the preset fusion model is less than a set threshold, thus obtaining the initial water... Fertilizer and water management decision model: The initial water and fertilizer decision model is reduced to different degrees to simplify the input and output data, resulting in the final water and fertilizer decision model. When the input data is the crop yield estimation coefficient, the corresponding output data is the target yield of the crop. When the input data is the crop nitrogen nutrition index, the corresponding output data is the relative yield of the crop. When the input data is the crop's suitable nitrogen accumulation, the corresponding output data is the crop's nitrogen supplementation regulation amount. When the input data is the nitrogen nutrition index, the corresponding output data is the nitrogen supplementation regulation amount. When the input data is the reference crop's water evapotranspiration, the corresponding output data is the crop's actual water consumption.

[0084] In one embodiment, the crop model fitting parameter determination module 201 is used to: measure the growth of crops to obtain multiple input data; input each input data into a water and fertilizer decision model to obtain the output data of the water and fertilizer decision model, and construct a data set based on each input data and the corresponding output data; construct transformation information of input data and corresponding output data based on the initial crop model fitting parameters; input multiple data sets into the transformation information to update the initial crop model fitting parameters and obtain the crop model fitting parameters.

[0085] In one embodiment, the preset parameters include nitrogen uptake per 100 kg of crop grain, basic soil nitrogen supply, basic nitrogen application, and nitrogen fertilizer utilization rate. The nitrogen topdressing regulation and irrigation amount determination module 202 is used to obtain the nitrogen topdressing regulation amount of the crop based on the crop model fitting parameters and preset parameters: updating the third transformation information based on the third crop model fitting parameters, and inputting the yield estimation coefficient into the updated third transformation information to update the target yield; or updating the fourth transformation information based on the fourth crop model fitting parameters, and inputting the nitrogen nutrient index into the updated fourth transformation information to update the relative yield, based on the relative yield and crop The target yield is updated by taking the extreme value of the maximum yield; the third crop model fitting parameters are the yield estimation coefficient and the crop model fitting parameters for the target yield; the fourth crop model fitting parameters are the nitrogen nutrient index and the crop model fitting parameters for the target yield; the third transformation information is the transformation information corresponding to the third crop model fitting parameters; the fourth transformation information is the transformation information corresponding to the fourth crop model fitting parameters; based on the updated target yield and the nitrogen uptake per 100 kg of crop grain, the nitrogen requirement of the crop is determined; based on the nitrogen requirement, the basic nitrogen supply of the soil, the basic nitrogen application, and the nitrogen fertilizer utilization rate, the amount of topdressing nitrogen regulation is determined.

[0086] In one embodiment, the preset parameters include the real-time soil volumetric water content of the crop, irrigation depth, soil moisture content, and the percentage of real-time soil volumetric water content. The nitrogen regulation and irrigation amount determination module 202 is used to obtain the irrigation amount based on the crop model fitting parameters and preset parameters: update the fifth transformation information based on the fifth crop model fitting parameters, input water evapotranspiration into the updated fifth transformation information to update the actual water consumption, the fifth crop model fitting parameters are the crop model fitting parameters of water evapotranspiration and actual water consumption, and the fifth transformation information is the transformation information corresponding to the fifth crop model fitting parameters; determine the irrigation amount based on the updated actual water consumption, real-time soil volumetric water content, irrigation depth, soil moisture content, and the percentage of real-time soil volumetric water content.

[0087] In one embodiment, the nitrogen topdressing regulation amount and irrigation amount determination module 202 is used to determine the nitrogen topdressing regulation amount: update the nitrogen nutrition index based on the fitting parameters of the first crop model, the first crop model fitting parameters being the crop model fitting parameters of the appropriate nitrogen accumulation amount and the nitrogen topdressing regulation amount; update the nitrogen topdressing regulation amount based on the fitting parameters of the second crop model and the updated nitrogen nutrition index, the second crop model fitting parameters being the crop model fitting parameters of the nitrogen nutrition index and the nitrogen topdressing regulation amount.

[0088] In one embodiment, the nitrogen regulation amount and irrigation amount determination module 202 is used for the nitrogen nutrition index: based on the aboveground biomass of the crop and the fitting parameters of the initial first crop model, the critical nitrogen concentration of the crop is determined; based on the ratio of the actual nitrogen concentration of the crop to the critical nitrogen concentration, the nitrogen nutrition index is obtained.

[0089] The figure is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 3 As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330, and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other via the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute a crop water and fertilizer decision-making method, which includes:

[0090] Based on the water and fertilizer decision-making model, crop model fitting parameters are obtained for crops. The water and fertilizer decision-making model is trained on the basis of the sample basic data and sample labels of sample crops, which are based on the pre-set fusion model. The sample basic data includes historical growth environment data and historical growth cycle data, and the sample labels include sample crop yield, sample nitrogen regulation amount, or sample water consumption. The nitrogen regulation amount and irrigation amount of crops are obtained based on the crop model fitting parameters and pre-set parameters. The water and fertilizer decision of crops is determined based on the nitrogen regulation amount and irrigation amount.

[0091] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent devices, 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 device. This computer software device 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.

[0092] In another aspect, 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 crop water and fertilizer decision determination method provided in the above embodiments, the method comprising:

[0093] Based on the water and fertilizer decision-making model, crop model fitting parameters are obtained for crops. The water and fertilizer decision-making model is trained on the basis of the sample basic data and sample labels of sample crops, which are based on the pre-set fusion model. The sample basic data includes historical growth environment data and historical growth cycle data, and the sample labels include sample crop yield, sample nitrogen regulation amount, or sample water consumption. The nitrogen regulation amount and irrigation amount of crops are obtained based on the crop model fitting parameters and pre-set parameters. The water and fertilizer decision of crops is determined based on the nitrogen regulation amount and irrigation amount.

[0094] 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.

[0095] 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 device. This computer software device 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.

[0096] 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 determining water and fertilizer application decisions for crops, characterized in that, include: Based on the water and fertilizer decision model, crop model fitting parameters of crops are obtained; wherein, the water and fertilizer decision model is trained on the basis of the sample basic data and sample labels of the sample crops, based on the preset fusion model. The sample basic data includes historical growth environment data and historical growth cycle data, and the sample labels include the sample yield, sample nitrogen regulation amount, or sample water consumption of the sample crops. The amount of nitrogen applied and the amount of irrigation for the crop are obtained based on the crop model fitting parameters and preset parameters. The water and fertilizer decision for the crop is determined based on the nitrogen supplementation regulation amount and the irrigation amount; The water and fertilizer decision model is obtained based on the following steps: The historical growth environment data is obtained based on meteorological data and soil basic data of the sample crops during their growth process. The historical growth cycle data is obtained based on crop canopy images, 3D point clouds, RGB images, spectral image data, thermal infrared data, and hyperspectral data during the growth process of the crops, so as to obtain the sample basic data. Based on the sample labels and the sample basic data, the sample basic data carrying the labels is obtained; The preset fusion model is trained and tested based on the sample data. When the error of the prediction result of the preset fusion model is less than a set threshold, the preset fusion model is determined to be trained and the initial water and fertilizer decision model is obtained. The initial water and fertilizer decision model is dimensionality reduced to different degrees to simplify the input and output data, resulting in the final water and fertilizer decision model. When the input data is the crop's yield estimation coefficient, the corresponding output data is the crop's target yield. When the input data is the crop's nitrogen nutrient index, the corresponding output data is the crop's relative yield. When the input data is the crop's suitable nitrogen accumulation, the corresponding output data is the crop's nitrogen topdressing regulation amount. When the input data is the nitrogen nutrient index, the corresponding output data is the nitrogen topdressing regulation amount. When the input data is the reference crop's water evapotranspiration, the corresponding output data is the crop's actual water consumption. The crop model fitting parameters for crops, based on the water and fertilizer decision-making model, include: The growth of the crops is measured to obtain multiple sets of input data; Each input data is input into the water and fertilizer decision model to obtain the output data of the water and fertilizer decision model, and a data group is constructed based on each input data and the corresponding output data. Based on the initial crop model fitting parameters, the transformation information of the input data and the corresponding output data is constructed. Multiple sets of the data are input into the transformation information to update the initial crop model fitting parameters, thereby obtaining the crop model fitting parameters.

2. The method for determining water and fertilizer application for crops according to claim 1, characterized in that, The preset parameters include nitrogen uptake per 100 kg of crop grain, basic soil nitrogen supply, basic nitrogen application rate, and nitrogen fertilizer utilization rate. Obtaining the nitrogen application rate for the crop based on the crop model fitting parameters and the preset parameters includes: The third transformation information is updated based on the fitting parameters of the third crop model, and the estimated yield coefficient is input into the updated third transformation information to update the target yield; or the fourth transformation information is updated based on the fitting parameters of the fourth crop model, and the nitrogen nutrient index is input into the updated fourth transformation information to update the relative yield, and the target yield is updated based on the extreme value of the relative yield and the maximum yield of the crop; the third crop model fitting parameters are the crop model fitting parameters of the estimated yield coefficient and the target yield, the fourth crop model fitting parameters are the crop model fitting parameters of the nitrogen nutrient index and the target yield, the third transformation information is the transformation information corresponding to the fitting parameters of the third crop model, and the fourth transformation information is the transformation information corresponding to the fitting parameters of the fourth crop model; Based on the updated target yield and the nitrogen uptake per 100 kg of grain of the crop, the nitrogen requirement of the crop is determined. The amount of nitrogen supplementation is determined based on the nitrogen requirement, the basic nitrogen supply to the soil, the basic nitrogen application rate, and the nitrogen fertilizer utilization rate.

3. The method for determining water and fertilizer application decisions for crops according to claim 1, characterized in that, The preset parameters include the real-time soil volumetric water content of the crop, irrigation depth, soil moisture content, and the percentage of real-time soil volumetric water content. The irrigation amount is obtained based on the crop model fitting parameters and the preset parameters, including: The fifth transformation information is updated based on the fifth crop model fitting parameters. The water evapotranspiration is input into the updated fifth transformation information to update the actual water consumption. The fifth crop model fitting parameters are the crop model fitting parameters of the water evapotranspiration and the actual water consumption. The fifth transformation information is the transformation information corresponding to the fifth crop model fitting parameters. The irrigation amount is determined based on the updated actual water consumption, the real-time soil volumetric water content, the irrigation depth, the soil moisture content, and the percentage of the real-time soil volumetric water content.

4. The method for determining water and fertilizer application for crops according to claim 1, characterized in that, Determining the nitrogen adjustment amount also includes: The nitrogen nutrition index is updated based on the fitting parameters of the first crop model, wherein the fitting parameters of the first crop model are the crop model fitting parameters of the appropriate nitrogen accumulation and the nitrogen application regulation amount. The nitrogen nutrient index is updated based on the fitting parameters of the second crop model and the updated nitrogen nutrient index. The fitting parameters of the second crop model are the crop model fitting parameters of the nitrogen nutrient index and the nitrogen nutrient index.

5. The method for determining water and fertilizer application for crops according to claim 4, characterized in that, Determine the nitrogen nutritional index: Based on the aboveground biomass of the crop and the fitting parameters of the initial first crop model, the critical nitrogen concentration of the crop is determined. The nitrogen nutrient index is obtained based on the ratio of the actual nitrogen concentration of the crop to the critical nitrogen concentration.

6. A device for determining water and fertilizer use for crops, characterized in that, include: The crop model fitting parameter determination module is used to obtain crop model fitting parameters for crops based on the water and fertilizer decision model. The water and fertilizer decision model is trained on the basis of the sample basic data and sample labels of the sample crops, based on the preset fusion model. The sample basic data includes historical growth environment data and historical growth cycle data. The sample labels include the sample yield, sample nitrogen regulation amount, or sample water consumption of the sample crops. The nitrogen application rate and irrigation amount determination module is used to obtain the nitrogen application rate and irrigation amount of the crop based on the crop model fitting parameters and preset parameters. A water and fertilizer decision determination module is used to determine the water and fertilizer decision for the crop based on the nitrogen application rate and the irrigation amount. The water and fertilizer decision model is obtained based on the following steps: Historical growth environment data is obtained based on meteorological and soil baseline data of the sample crops during their growth process; historical growth cycle data is obtained based on crop canopy images, 3D point clouds, RGB images, spectral image data, thermal infrared data, and hyperspectral data during the crop's growth process, to obtain the sample baseline data; Labeled sample baseline data is obtained based on the sample labels and the sample baseline data; the preset fusion model is trained and tested based on the sample baseline data; when the error of the prediction result of the preset fusion model is less than a set threshold, the preset fusion model is determined to be trained successfully, and an initial water and fertilizer decision model is obtained; the initial water and fertilizer decision model is then... The decision model undergoes dimensionality reduction to varying degrees to simplify the input and output data of the initial water and fertilizer decision model, resulting in the water and fertilizer decision model. When the input data is the crop's yield estimation coefficient, the corresponding output data is the crop's target yield. When the input data is the crop's nitrogen nutrient index, the corresponding output data is the crop's relative yield. When the input data is the crop's suitable nitrogen accumulation, the corresponding output data is the crop's nitrogen topdressing regulation amount. When the input data is the nitrogen nutrient index, the corresponding output data is the nitrogen topdressing regulation amount. When the input data is the reference crop's water evapotranspiration, the corresponding output data is the crop's actual water consumption. The crop model fitting parameter determination module is used to measure the growth of the crop and obtain multiple input data; input each input data into the water and fertilizer decision model to obtain the output data of the water and fertilizer decision model; construct a data group based on each input data and the corresponding output data; construct transformation information of the input data and the corresponding output data based on the initial crop model fitting parameters; input multiple data groups into the transformation information to update the initial crop model fitting parameters and obtain the crop model fitting parameters.

7. 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 crop water and fertilizer decision determination method as described in any one of claims 1 to 5.

8. 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 crop water and fertilizer decision determination method as described in any one of claims 1 to 5.

Images