Facility vegetable demand-driven water and fertilizer intelligent management and control method and system

Abstract

The application provides a water and fertilizer intelligent management and control method and system based on facility vegetable demand driving, and relates to the technical field of intelligent agriculture. Based on meteorological data and crop target yield, the total nutrient demand amount of the vegetable crop in each growth period is dynamically predicted through a crop growth model, the water and fertilizer management and control are driven by crop demand, the water and fertilizer supply is matched with the actual growth demand of the crop, the nutrient recommendation coefficient is determined by introducing soil fertility grade analysis, the accurate check based on the actual soil fertilization capacity is realized, and the rationality of nutrient input is significantly improved. Further, the initial scheme is formulated by combining the organic and inorganic fertilizer matching principle and the organic fertilizer mineralization characteristics, and the scientificity and sustainability of nutrient supply are ensured. Finally, the execution scheme is generated through the multi-objective optimization of the decision support system, a complete decision closed loop from prediction, planning to optimization is formed, and the accuracy and reliability of the water and fertilizer management and control of the facility vegetable are improved.

Description

Technical Field

[0001] This invention relates to the field of smart agriculture technology, and in particular to a smart water and fertilizer management method and system based on the demand-driven nature of greenhouse vegetables. Background Technology

[0002] As an important component of modern agriculture, the precision and intelligence of water and fertilizer management in greenhouse vegetable production directly affect crop yield, quality, and resource utilization efficiency. In recent years, decision support technology based on crop growth models has been widely used in the field of water and fertilizer management. By simulating crop physiological and ecological processes, it predicts water and fertilizer requirements, providing a theoretical basis for production management.

[0003] However, most current water and fertilizer decision-making methods are static, primarily based on historical data or empirical models to generate water and fertilizer plans, lacking dynamic response capabilities. For example, when actual environmental conditions deviate from the model's preset conditions, such static plans are difficult to adapt to the real-time needs of crops, easily leading to a mismatch between water and fertilizer supply and crop growth, resulting in resource waste and even crop physiological stress.

[0004] Furthermore, current water and fertilizer decision-making methods suffer from model dependence and monitoring blind spots. The systems rely excessively on idealized physical mechanism models for prediction, lacking effective coordination with precision monitoring modules. When faced with sudden stresses, such as pest and disease outbreaks, extreme high temperatures, or root diseases, they cannot capture abnormal changes in crop growth status through real-time sensing mechanisms, leading to delayed system response and significantly reduced decision robustness.

[0005] Therefore, current water and fertilizer management technologies for greenhouse vegetables suffer from insufficient responsiveness and poor environmental adaptability. Summary of the Invention

[0006] This invention provides a smart water and fertilizer management method and system based on the demand-driven nature of greenhouse vegetables, which solves the problems of insufficient response and poor environmental adaptability in the current water and fertilizer management technology for greenhouse vegetables, and improves the accuracy and reliability of water and fertilizer management for greenhouse vegetables.

[0007] Firstly, this invention provides a smart water and fertilizer management method driven by the demand of greenhouse vegetables. The method includes: acquiring greenhouse meteorological data, soil data, and crop target data for the crop to be fertilized; predicting the total nutrient demand of the crop at each growth stage based on the greenhouse meteorological data, crop target data, and a pre-set greenhouse vegetable growth model; analyzing soil fertility levels based on soil data and determining nutrient recommendation coefficients; determining the total nutrient input based on the total nutrient demand and nutrient recommendation coefficients; determining an initial water and fertilizer scheme for organic and inorganic fertilizers based on the total nutrient input, the principle of organic-inorganic fertilizer application, and the mineralization characteristics of organic fertilizer nutrients; and performing multi-objective rolling decision optimization based on the initial water and fertilizer scheme, crop target data, and a decision support system to generate the final water and fertilizer scheme.

[0008] Secondly, embodiments of the present invention provide a smart water and fertilizer management device driven by the needs of greenhouse vegetables. The device includes a communication module and a processing module. The communication module is used to acquire greenhouse meteorological data, soil data, and crop target data of the crop to be fertilized. The processing module is used to predict the total nutrient requirements of the crop at each growth stage based on the greenhouse meteorological data, crop target data, and a preset greenhouse vegetable growth model; analyze soil fertility levels based on soil data and determine nutrient recommendation coefficients; determine the total nutrient input based on the total nutrient requirements and nutrient recommendation coefficients; determine the initial water and fertilizer scheme for organic and inorganic fertilizers based on the total nutrient input, the principle of organic-inorganic fertilizer application, and the mineralization characteristics of organic fertilizer nutrients; and perform multi-objective rolling decision optimization based on the initial water and fertilizer scheme, crop target data, and a decision support system to generate the final water and fertilizer scheme.

[0009] Thirdly, embodiments of the present invention provide a decision support system, which includes an electronic device. The electronic device includes a memory and a processor. The memory stores a computer program, and the processor is used to call and run the computer program stored in the memory to perform the steps of the method as described in the first aspect and any possible implementation thereof.

[0010] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, it implements the steps of the method as described in the first aspect and any possible implementation thereof.

[0011] This invention provides a smart water and fertilizer management method and system based on the demand-driven nature of greenhouse vegetables. Based on meteorological data and target crop yield, the invention dynamically predicts the total nutrient requirements of vegetable crops at each growth stage using a crop growth model. Water and fertilizer management is driven by crop demand, ensuring that water and fertilizer supply matches the actual growth needs of the crop. By introducing soil fertility level analysis to determine nutrient recommendation coefficients, precise verification based on the actual soil fertility capacity is achieved, significantly improving the rationality of nutrient input. Furthermore, by combining the principles of organic and inorganic fertilizer application with the mineralization characteristics of organic fertilizer, an initial plan is formulated, ensuring the scientific and sustainable nature of nutrient supply. Finally, a decision support system is used to perform multi-objective optimization to generate an execution plan, forming a complete decision-making closed loop from prediction and planning to optimization. This solves the problems of insufficient responsiveness and poor environmental adaptability in current water and fertilizer management technologies for greenhouse vegetables, improving the accuracy and reliability of water and fertilizer management for greenhouse vegetables. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart illustrating a smart water and fertilizer management method based on demand-driven greenhouse vegetable production, as provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of a smart water and fertilizer management device based on the demand-driven nature of greenhouse vegetables, provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0014] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0015] In the description of this invention, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The term "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" and "more than one" refer to two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences.

[0016] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner to facilitate understanding.

[0017] Furthermore, the terms "comprising" and "having," and any variations thereof, used in the description of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or modules is not limited to the steps or modules listed, but may optionally include other steps or modules not listed, or may optionally include other steps or modules inherent to such process, method, product, or device.

[0018] To make the objectives, technical solutions, and advantages of the present invention clearer, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0019] like Figure 1 As shown, this embodiment of the invention provides a smart water and fertilizer management method based on the demand-driven nature of greenhouse vegetables. The method includes steps S101-S106.

[0020] S101. Obtain facility meteorological data, soil data, and crop target data for the crops to be fertilized.

[0021] In some embodiments, facility meteorological data includes light intensity, temperature and humidity, soil data includes soil fertility level, initial soil nutrient content and soil moisture data, and crop target data includes target yield and / or target quality.

[0022] In some embodiments, soil data include organic matter content, available nitrogen content, available phosphorus content, available potassium content, pH value, soil bulk density, and soil texture; For example, embodiments of the present invention can continuously collect facility environmental parameters, including light intensity, air temperature, and air humidity, through IoT sensing devices deployed in greenhouses or polytunnels; simultaneously obtain basic soil data such as soil organic matter, nitrogen, phosphorus, and potassium content, and pH value through soil testing; and receive crop target data set by the user, including target yield and fruit quality requirements.

[0023] S102. Based on facility meteorological data and crop target data, as well as a pre-set facility vegetable growth model, predict the total nutrient requirements of the crop to be fertilized at each growth stage.

[0024] In some embodiments, greenhouse vegetables include tomatoes, cucumbers, eggplants, and bell peppers.

[0025] In some embodiments, greenhouse vegetables also include melons such as watermelon, cantaloupe, pumpkin, and zucchini, and climbing vines such as green beans and cowpeas.

[0026] In some embodiments, the present invention can input real-time collected environmental data and crop target data into a validated greenhouse vegetable growth model. This model accurately predicts the total nitrogen, phosphorus, and potassium nutrients required to achieve the target yield during each growth stage from transplanting to harvest by simulating the photosynthetic efficiency, dry matter accumulation, and distribution processes of crops under different environmental conditions. This prediction fully considers the dynamic impact of environmental factors on crop growth and nutrient requirements.

[0027] As one possible implementation, step S102 can be specifically implemented as steps S1021-S1024.

[0028] S1021. Using facility meteorological data as the core driving variable, input the model of facility vegetable growth to simulate the photosynthetic process and dry matter accumulation process of each growth stage of the crop to be fertilized from transplanting to harvest.

[0029] For example, in embodiments of the present invention, real-time monitored light and temperature data can be used as core driving variables and input into a greenhouse vegetable growth model. The model dynamically simulates the photosynthetic rate of crops at various key growth stages (such as seedling stage, flowering and fruit setting stage, and fruit enlargement stage) and the resulting dry matter formation and accumulation process through built-in physiological and ecological algorithms.

[0030] In some embodiments, the greenhouse vegetable growth model is a heat-time driven mechanism-data fusion growth model. It takes cumulative heat time as the core development timeline, couples meteorological data such as solar radiation and temperature, dynamically analyzes the transformation relationship between crop growth and development and radiation and accumulated temperature, accurately estimates the amount of dry matter accumulation and derives nutrient requirements. It can be adapted to greenhouse vegetables such as tomatoes, cucumbers, and cowpeas, and has made growth cycle mutation corrections for topping agronomic measures, solving the problem of traditional models overestimating dry matter accumulation and nutrient absorption.

[0031] For example, the greenhouse vegetable growth model is a multi-module coupled dynamic prediction architecture, including five core modules: thermal time-driven module, photosynthetic production module, dry matter accumulation and distribution module, nutrient requirement derivation module, and topping correction module. Each module dynamically iterates with a daily time step. The inputs are meteorological data (solar radiation, temperature, air humidity), crop target yield, and basic crop parameters. The outputs are the dry matter accumulation, total nitrogen, phosphorus, and potassium nutrient requirements, and absorption intensity at each growth stage. The coupling relationship between modules is: thermal time-driven → photosynthetic production → dry matter accumulation → topping correction → nutrient requirement derivation.

[0032] For example, the thermal time-driven module is a time-series reference module that undertakes the calculation logic of actual cumulative thermal time and target cumulative thermal time, converts temperature data into thermal time indicators that characterize the physiological development process of crops, and calculates growth stage parameters (values ​​0-1).

[0033] Actual cumulative heating time: ; Growth stage parameters: .

[0034] Among them, T act T represents the actual cumulative thermal time. tar For the target cumulative thermal time, T avg-i Let T be the average daily temperature on day i. b The base temperature for crop growth is S=1, which represents the developmental process (such as the topping node) corresponding to the crop reaching the target yield.

[0035] For example, the photosynthetic production module uses solar radiation data as the core energy input to calculate the actual photosynthetically active radiation and radiation utilization rate available to crops, providing a basis for dry matter accumulation.

[0036] Photosynthetically active radiation: PAR = Rs × 0.45, where Rs is the total daily solar radiation (MJ / m2) and 0.45 is the measured value of the proportion of photosynthetically active radiation in greenhouse vegetables.

[0037] Canopy radiation interception rate: f=1 e k×LAI Where k is the extinction coefficient (0.7 for tomatoes, 0.65 for cucumbers, measured values ​​for greenhouse vegetables), and LAI is the leaf area index, dynamically fitted from the actual cumulative heat time. Among them, LAI max Maximum leaf area index (MAI): The upper limit of leaf area index that a crop can reach when it reaches its vigorous growth stage.

[0038] Dynamic radiation utilization rate: RUE = RUE max ×fT ×f W Among them, RUE max For maximum radiation utilization (average 2.8 g / MJ for greenhouse vegetables), f T For temperature correction factor, f W This is a humidity correction factor, which can be dynamically calculated from meteorological data or based on experience. For example, the temperature correction factor f... T (Suitable for greenhouse vegetable growth) Value range: 0~1. Suggested value rules: Suitable temperature (15℃-30℃): 0.2~1.0 (maximum value of 1.0 at 25℃, linearly decreasing towards both ends); Unsuitable temperature (<15℃ or ≥30℃): 0~0.2 (0 for <10℃ / ≥35℃). Humidity correction factor f W (Suitable for greenhouse vegetable growth) Value range: 0~1. Recommended value rules: Suitable humidity (40%-90%): 0.3~1.0 (75% takes the maximum value of 1.0, and decreases linearly towards both ends); Unsuitable humidity (<40% or >90%): 0~0.3 (<30% / >95% takes 0).

[0039] For example, the dry matter accumulation and distribution module calculates the theoretical dry matter accumulation based on the output of the photosynthetic production module, and performs dry matter distribution among organs in combination with the crop growth stage.

[0040] Theoretical dry matter accumulation: W theo =PAR×f×RUE×0.95, where 0.95 is the efficiency of photosynthetic products in converting into dry matter.

[0041] Actual dry matter accumulation: W act =W theo ×S, through the correction of the growth stage parameter S, eliminates the problem of overestimation of dry matter accumulation caused by topping.

[0042] Organ dry matter allocation: Based on the actual cumulative heat time, the growth stage is determined, and the organ allocation coefficient database is called up (e.g., during the flowering and fruit setting period of tomatoes: fruit 30%, stems and leaves 60%, roots 10%) to complete the allocation of dry matter to each organ.

[0043] For example, the nutrient requirement derivation module, coupled with the nutrient dilution curve and nutrient ratio relationship of this application, derives the nitrogen, phosphorus and potassium nutrient requirements from the cumulative dry matter, and is the core output module.

[0044] Nitrogen concentration: determined by crop-specific nutrient dilution curves, such as for tomatoes: nitrogen concentration ; Nitrogen demand: N=W act ×N% Phosphorus / potassium requirements: Calculated based on the measured ratio of nitrogen to phosphorus (1:0.25) and nitrogen to potassium (1:1.2), i.e., phosphorus requirement P=N×0.25 and potassium requirement K=N×1.2.

[0045] For example, the topping correction module uses the growth stage parameter S and the target cumulative thermal time T. tar Constraints on the growth process after topping: When T act ≥T tar When S=1, the accumulation of dry matter stops growing, and the nutrient demand no longer increases. This accurately matches the growth pattern of crops after topping in actual production, and solves the deviation between the ideal assumption of unlimited growth in traditional models and actual production.

[0046] In some embodiments, the training data for the greenhouse vegetable growth model comes from: the experiment was conducted at a greenhouse vegetable experimental base, with tomatoes, cucumbers, and cowpeas as the test crops. Three target yield gradients and five transplanting time gradients were set. Core indicators such as cumulative heat time, leaf area index, dry matter accumulation, and nutrient uptake were measured under each treatment. At the same time, meteorological data (solar radiation, temperature, and humidity) were collected throughout the entire growth period, and a total of 1,500 sets of valid sample data were obtained, covering the main cultivation scenarios of greenhouse vegetables.

[0047] This invention uses the least squares method to fit and calibrate key parameters in the model (such as extinction coefficient k, maximum radiation utilization rate RUE, nutrient dilution curve parameters, and organ allocation coefficient), with the actual dry matter accumulation and actual nutrient absorption as target values. Through iterative calculation, the root mean square error (RMSE) between the model prediction and the measured value is less than 6%.

[0048] This invention employs both independent field trials and on-site measurements at production bases for dual verification. A control group was set up with topping and no topping. The correlation coefficients (R²) between the model's predicted dry matter accumulation and nutrient requirements and the measured values ​​were all greater than 0.93, and the prediction errors for nitrogen, phosphorus, and potassium nutrient requirements were all less than 8%, verifying the accuracy and practicality of the model under topping agronomic measures.

[0049] S1022. During the simulation, based on the target crop data, the biomass distribution ratio of each organ of the crop to be fertilized is determined, and the total amount of dry matter required by the crop to be fertilized is calculated.

[0050] For example, embodiments of the present invention can, based on the target yield set by the user, and according to the known harvest index (the ratio of economic yield to biomass) of the crop variety and the standard biomass distribution ratio of each organ (fruit, stems and leaves, roots), reverse-calculate the total dry matter required to achieve the target yield.

[0051] S1023. Based on the nutrient absorption patterns of the crop to be fertilized and the dry matter accumulation process at each growth stage, calculate the amount and intensity of nitrogen, phosphorus, and potassium nutrients absorbed at each growth stage.

[0052] For example, embodiments of the present invention can calculate the amount of nitrogen, phosphorus, and potassium required for each growth stage based on the dynamic curves of dry matter accumulation obtained from the above simulation, combined with the established laws in crop nutrition regarding the intensity and ratio of nitrogen, phosphorus, and potassium absorption at different growth stages, and clarify their absorption intensity.

[0053] S1024. Calculate the total nutrient requirement based on the total dry matter required by the crop to be fertilized, as well as the absorption amount and intensity of nitrogen, phosphorus, and potassium nutrients at each growth stage.

[0054] In some embodiments, the total nutrient requirement includes the total nitrogen requirement, the total phosphorus requirement, and the total potassium requirement.

[0055] For example, embodiments of the present invention can sum up the nitrogen, phosphorus, and potassium absorption calculated at all growth stages throughout the entire growth period to finally obtain the total nitrogen, phosphorus, and potassium requirements necessary to achieve the set target yield and that match the dynamic growth process.

[0056] S103. Based on soil data, analyze soil fertility levels and determine nutrient recommendation coefficients.

[0057] In some embodiments, the present invention can objectively classify soil fertility into high, medium, and low levels based on multiple indicators obtained from soil testing and using comprehensive soil fertility evaluation methods such as weighted scoring. According to the inherent nutrient supply and retention capacity of soils at different fertility levels, an appropriate nutrient recommendation coefficient is determined. This coefficient is used to correct theoretical fertilization amounts, achieving soil-specific fertilization.

[0058] As one possible implementation, step S103 can be specifically implemented as steps S1031-S1032.

[0059] S1031. Based on soil data, the soil fertility level is determined by using the comprehensive soil fertility evaluation method.

[0060] In some embodiments, soil fertility grades include high fertility, medium fertility, and low fertility.

[0061] For example, embodiments of the present invention can obtain multiple key indicator data through standardized soil testing methods. Among them: organic matter content refers to the total amount of organic matter in the soil, which is a core indicator for evaluating the overall level of soil fertility; alkaline available nitrogen content reflects the level of nitrogen that the soil can supply for crop absorption and utilization in the near future; available phosphorus content characterizes the phosphorus component in the soil that can be absorbed by the current season's crops; available potassium content indicates the content of potassium element in the soil that is easily absorbed by crops; pH value represents the soil acidity and alkalinity, which directly affects nutrient availability and the crop root growth environment; soil bulk density indicates the actual water content per unit volume of soil; soil texture refers to the relative proportion of mineral particles of different sizes in the soil, which affects water and fertilizer retention capacity.

[0062] For example, this embodiment of the invention uses a comprehensive soil fertility evaluation method to systematically analyze the above-mentioned indicators. Specifically, a standardized scoring system for each indicator is established, and corresponding weights are assigned based on the degree of influence of different indicators on crop growth; the measured soil data is compared with local soil fertility grading standards, and each indicator is scored independently; a comprehensive soil fertility score is obtained through weighted calculation; based on the comprehensive score, soil fertility is divided into three distinct levels: high, medium, and low. High-fertility soils have balanced and excellent indicators; medium-fertility soils have moderate levels of main indicators; and low-fertility soils exhibit significant nutrient deficiencies or physical and chemical property obstacles.

[0063] S1032. Based on the soil fertility level and the pre-established mapping relationship between soil fertility level and nutrient recommendation coefficient, determine the nutrient recommendation coefficient.

[0064] For example, in this embodiment of the invention, based on a determined soil fertility level, a pre-established database mapping soil fertility level to nutrient recommendation coefficients is invoked: High-fertility soils, due to their strong basic nutrient supply capacity, are assigned a lower nutrient recommendation coefficient (0.7-0.9) to fully utilize their own nutrients; medium-fertility soils are assigned a moderate recommendation coefficient (around 1.0) to maintain a balance between nutrient supply and demand; and low-fertility soils, due to insufficient nutrient supply capacity, are assigned a higher recommendation coefficient (1.1-1.3) to supplement soil nutrient deficiencies through fertilization. This process ensures that nutrient recommendations meet crop needs while fully considering the actual nutrient supply capacity of the soil, providing key parameter basis for subsequent precision fertilization.

[0065] S104. Determine the total amount of nutrient input based on the total nutrient requirement and the nutrient recommendation coefficient.

[0066] In some embodiments, the present invention combines the total theoretical nutrient requirement predicted by the growth model with a recommendation coefficient based on soil fertility, and calculates through mathematical calculations a total amount of scientific nutrient input that meets the requirements for high-yield and high-quality crops while fully considering the soil's own nutrient supply capacity.

[0067] As one possible implementation, step S104 can be specifically implemented as steps S1041-S1045.

[0068] S1041. Based on the total nutrient demand and the dynamic nutrient allocation principle based on the crop growth stage, the time period is divided according to the physiological demand characteristics of different growth stages in each growth period, and multiple growth stages are determined, as well as the nutrient demand allocation of each growth stage.

[0069] For example, embodiments of the present invention can divide each growth period into multiple key growth stages based on the dynamic nutrient allocation principle of crop growth stages, including but not limited to the seedling stage, flowering and fruit setting stage, fruit enlargement stage, and maturity stage. The corresponding nutrient requirements are determined according to the different physiological characteristics and growth centers of each stage. In specific implementation, based on the dry matter accumulation curve output by the crop growth model and the organ formation patterns, the predicted total nutrient requirements for each growth period are proportionally allocated to each growth stage, forming a stage-based nutrient requirement allocation table.

[0070] S1042. Based on soil fertility level, the nutrient recommendation coefficient is modified specifically for each growth stage to generate dynamic recommendation coefficients for each growth stage.

[0071] For example, embodiments of the present invention perform stage-specific adjustments to a uniform nutrient recommendation coefficient based on a determined soil fertility level. Specifically, dynamic recommendation coefficients are generated for each growth stage, taking into account differences in crop root development, nutrient absorption capacity, and soil environmental changes at different growth stages. For instance, in the early stages of crop growth, when root distribution is shallow and absorption capacity is weak, the recommendation coefficient is appropriately increased; during the peak growth period, when crop absorption capacity is strong, a standard or appropriately reduced recommendation coefficient is used.

[0072] S1043. The nutrient demand allocation and dynamic recommendation coefficient of each reproductive stage are weighted and integrated to calculate the dynamic nutrient input of each reproductive stage.

[0073] For example, embodiments of the present invention can perform weighted fusion calculations of nutrient requirements at each growth stage and corresponding dynamic recommendation coefficients. In specific implementation, a weighted algorithm considering the importance of each stage is used, assigning higher weights to key crop growth periods, such as the fruit enlargement stage, to ensure nutrient supply during critical periods. By calculating and accumulating step-by-step, the dynamic nutrient input for each growth stage is obtained, and this value fully reflects the dynamic matching between crop stage requirements and soil supply characteristics.

[0074] S1044. Based on facility meteorological data, generate environmental factors.

[0075] In some embodiments, environmental factors are used to characterize the extent to which facility meteorological data affects the total amount of nutrient input.

[0076] For example, embodiments of the present invention can construct a comprehensive environmental factor based on real-time acquired facility meteorological data. This environmental factor is a comprehensive index that quantifies the impact of environmental conditions on nutrient utilization efficiency. By integrating monitoring data of environmental parameters such as light, temperature, and humidity, the actual nutrient utilization efficiency of crops under current environmental conditions is assessed. In specific implementation, a response relationship model between environmental parameters and nutrient efficiency is established. When environmental conditions are favorable for nutrient absorption, the nutrient input is appropriately reduced; conversely, under unfavorable environmental conditions, the nutrient input is maintained or appropriately increased to ensure that crops can obtain suitable nutrient supply under different environmental conditions.

[0077] S1045. Based on environmental factors, the dynamic nutrient input at each growth stage is corrected to obtain the total nutrient input.

[0078] For example, embodiments of the present invention can use environmental factors as correction coefficients to finally adjust the dynamic nutrient input at each growth stage. This results in a total nutrient input that not only considers the needs of the crop stage and the soil's supply capacity but also incorporates the influence of actual environmental conditions, forming a scientific total fertilization amount based on the comprehensive optimization of the crop-soil-environment system.

[0079] S105. Based on the total amount of nutrient input and the principle of applying organic and inorganic fertilizers in combination, and combined with the mineralization characteristics of organic fertilizer nutrients, determine the initial water and fertilizer scheme for organic and inorganic fertilizers.

[0080] In some embodiments, the present invention can use the total amount of nutrient input determined in the previous step as a constraint, follow the agricultural principle of applying organic fertilizers and inorganic fertilizers in combination, and comprehensively consider the differences in the nutrient release rate and pattern of different types of organic fertilizers (such as manure, straw, etc.) to formulate an initial water and fertilizer scheme that includes specific fertilizer types, dosages, ratios and application times of base fertilizer and topdressing.

[0081] As one possible implementation, step S105 can be specifically implemented as steps S1051-S1057.

[0082] S1051. Based on the total nutrient input, allocate nutrients according to the preset ratio of base fertilizer to topdressing to determine the target nutrient supply for the base fertilizer stage.

[0083] For example, embodiments of the present invention can rationally allocate the total nutrient input for each growth stage calculated in the aforementioned steps according to a preset ratio of basal fertilizer to topdressing. This allocation principle comprehensively considers the crop root development pattern and nutrient demand characteristics: in the early growth stage, the crop root system has a small distribution range and weak absorption capacity, so basal fertilizer needs to provide a stable and continuous supply of nutrients; while in the middle and late growth stages, topdressing meets the needs of rapid crop growth. Specifically, based on the growth characteristics of different vegetable crops, the basal fertilizer ratio is usually set between 40% and 60% of the total nutrient input, thereby accurately calculating the target nutrient quantity that needs to be supplied during the basal fertilizer stage.

[0084] S1052. Based on the target nutrient supply at the basal fertilizer stage and the principle of organic-inorganic fertilizer application, analyze the nutrient ratio that organic fertilizer should bear in the basal fertilizer and determine the target fertilizer supply of organic fertilizer.

[0085] For example, embodiments of the present invention can determine the proportion of nutrients that organic fertilizer should provide in the base fertilizer based on the principle of combined application of organic and inorganic fertilizers. This fully leverages the advantages of organic fertilizers in improving soil and providing comprehensive nutrition, while utilizing the high nutrient concentration and rapid effectiveness of inorganic fertilizers to achieve complementary benefits. In specific implementation, based on the basic soil fertility status and crop requirements, the organic fertilizer is set to provide 20%-50% of the nutrients during the base fertilizer stage, and the target nutrient supply required by the organic fertilizer is calculated accordingly.

[0086] S1053. Based on the target fertilizer supply of organic fertilizer, as well as the nutrient content, carbon-nitrogen ratio, decomposition degree and mineralization rate characteristics of various organic fertilizers, multiple candidate organic fertilizer varieties were screened from available organic fertilizer resources.

[0087] For example, embodiments of the present invention can select suitable varieties from locally available organic fertilizer resources based on the target fertilizer supply amount. The screening process comprehensively considers the following key indicators: nutrient content: the amount of nutrients such as nitrogen, phosphorus, and potassium contained in a unit weight of organic fertilizer; carbon-nitrogen ratio: the ratio of carbon to nitrogen in organic fertilizer, which affects the decomposition rate of organic fertilizer in the soil; maturity: reflecting the degree of fermentation maturity of organic fertilizer, directly affecting application safety and nutrient availability; mineralization rate characteristics: describing the rate and duration of nutrient release from organic fertilizer in the soil. By establishing a multi-indicator evaluation system, 3-5 candidate varieties with excellent overall performance are selected from the candidate organic fertilizers for further analysis.

[0088] S1054. Based on multiple candidate organic fertilizer varieties and facility meteorological data, combined with the mineralization characteristics of organic fertilizer nutrients, calculate the effective nutrient release amount of each candidate organic fertilizer variety during the basal fertilizer period.

[0089] For example, embodiments of the present invention can predict the effective nutrient release amount during the basal fertilizer period for each selected candidate organic fertilizer variety by combining real-time monitoring data of the facility environment. In specific implementation, an organic fertilizer mineralization kinetic model is established. This model comprehensively considers the influence of environmental factors such as soil temperature, humidity, and pH value on the organic fertilizer decomposition process, and simulates and calculates the effective nutrients that the organic fertilizer can actually release and be absorbed by crops from application to the end of the basal fertilizer effect.

[0090] In some embodiments, the organic fertilizer mineralization kinetic model is a first-order kinetic correction model driven by both thermal time and environmental factors, coupled with temperature data and facility soil environmental parameters (humidity, pH value), accurately simulating the effective release law of nitrogen, phosphorus and potassium of organic fertilizer under facility cultivation conditions, outputting the effective release of nutrients by organic fertilizer during the basal fertilizer period, providing a core basis for determining the total amount of nutrient input and formulating organic and inorganic fertilizer application schemes, and is compatible with commonly used organic fertilizers for facility vegetables such as decomposed chicken manure, cow manure, sheep manure, and straw compost.

[0091] In some embodiments, the organic fertilizer mineralization kinetic model is a dual-driven mineralization prediction architecture, including three major modules: a thermal time mineralization rate correction module, an organic fertilizer basic mineralization module, and a soil environmental factor correction module. Each module iterates synchronously with the actual cumulative thermal time of the facility vegetable growth model, with a daily time step. The inputs are organic fertilizer type / application amount, soil temperature / humidity / pH value, and crop planting date (the starting point of thermal time). The outputs are the daily effective release amount and total effective release amount of organic fertilizer nitrogen, phosphorus, and potassium during the basal fertilizer period (0-60 days).

[0092] For example, the thermal time mineralization rate correction module takes temperature data and actual cumulative thermal time calculation logic, converts soil temperature into mineralization thermal time, corrects the organic fertilizer mineralization rate, and realizes the synchronization of the mineralization process with the timing of crop growth and development.

[0093] Average daily soil temperature: T s =(T s max +T s min ) / 2, measured by the facility's soil temperature sensor; T s max The highest temperature of the day, T s min This is the lowest temperature of the day.

[0094] Effective thermal time for mineralization: (T) s-i ≥10, otherwise 0), 10℃ is the basic temperature for organic fertilizer mineralization. s i Let be the average daily soil temperature (°C) on day i. Thermal time mineralization correction factor: f Ts =e 0.05×(T s act / 10 5) This allows for dynamic adjustment of the mineralization rate based on crop growth heat time.

[0095] For example, the organic fertilizer basic mineralization module describes the mineralization law of organic fertilizer under standard conditions (soil temperature 25℃, humidity 70% field water holding capacity, pH 7.0) based on the first-order kinetic equation, and determines the basic mineralization parameters of different organic fertilizers.

[0096] M t =M0×(1 e k×t ) Among them, M t Let M0 be the effective nutrient release at time t (g / kg), M0 be the mineralizable nutrient release (g / kg), and k be the basic mineralization rate constant (d). - ¹), where t is the mineralization time (d). The basic mineralization parameters of commonly used organic fertilizers for greenhouse vegetables are determined by field measurements.

[0097] For example, the soil environmental factor correction module introduces soil moisture and pH correction coefficients to perform secondary correction on the basic mineralization rate constant, thereby achieving the matching of the mineralization process with the actual soil environment of the facility.

[0098] Humidity correction factor: W represents the actual soil moisture, applicable range 40%-90%; pH correction factor: f pH =1 0.05 × |pH 7.0 | Applicable pH range 5.5-8.5 (suitable pH for greenhouse vegetables); Actual mineralization rate constant: k′=k×f Ts ×f W ×f pH It integrates thermal time, humidity, and pH value for comprehensive correction.

[0099] In some embodiments, the present invention adopts a combination of indoor constant temperature culture and field in-situ monitoring, setting 5 soil temperature gradients (10, 15, 20, 25, 30℃), 4 humidity gradients (40%, 60%, 70%, 90%), and 4 pH gradients (5.5, 6.5, 7.0, 8.5). The tested organic fertilizer is a commonly used variety in facilities, the culture period is 60 days (the effective action period of the basal fertilizer), the available nitrogen, phosphorus and potassium content in the soil is measured regularly, the mineralization release of organic fertilizer is calculated, and the mineralization heat time is recorded simultaneously, accumulating 1200 sets of valid sample data.

[0100] This invention uses multiple linear regression to fit the calculation formulas of thermal time mineralization correction coefficient and environmental factor correction coefficient, taking the measured effective nutrient release as the target value, and through iterative calculation to make the RMSE between the model prediction value and the measured value less than 7%.

[0101] Field validation was conducted in greenhouse tomato / cucumber fields. Different organic fertilizers were applied and the available nutrient content in the soil was monitored during the basal fertilizer period. The R² of the model's predicted effective release of organic fertilizer was greater than 0.90, and the prediction error of nitrogen, phosphorus and potassium mineralization release was less than 9%, which verified the applicability of the model under actual greenhouse cultivation conditions.

[0102] S1055. Compare the target fertilizer supply amount of organic fertilizer with the effective nutrient release amount of each candidate organic fertilizer variety to generate the corresponding inorganic fertilizer supplement amount for each candidate organic fertilizer variety.

[0103] For example, embodiments of the present invention can compare and analyze the predicted effective nutrient release amounts of each candidate organic fertilizer variety with the target nutrient supply amount of the organic fertilizer. When the effective nutrient release amount of the organic fertilizer is lower than the target nutrient supply amount, the difference is calculated as the amount of nutrients that need to be supplemented by inorganic fertilizer; when the organic fertilizer release amount is sufficient or excessive, the amount of inorganic fertilizer supplemented is adjusted accordingly. This process ensures that the total nutrient supply during the basal fertilizer stage meets the crop's needs while avoiding excessive fertilization.

[0104] S1056. Based on the inorganic fertilizer supplementation amount corresponding to each candidate organic fertilizer variety and the mineralization rate characteristics of each candidate organic fertilizer variety, inorganic fertilizer is screened to obtain inorganic fertilizer varieties that match each candidate organic fertilizer variety.

[0105] For example, embodiments of the present invention can screen suitable fast-acting inorganic fertilizer varieties based on the inorganic fertilizer supplementation amount corresponding to each candidate organic fertilizer scheme and the mineralization rate characteristics of each organic fertilizer. The screening principle focuses on the matching of nutrient release rates: for organic fertilizers with slower mineralization rates, fast-acting inorganic fertilizers are applied in combination; for organic fertilizers with faster mineralization rates, slow-release inorganic fertilizers are selected, thereby achieving effective connection in the release sequence of organic and inorganic nutrients and ensuring that crops receive a balanced nutrient supply throughout the entire basal fertilizer period.

[0106] S1057. Based on each candidate organic fertilizer variety, the corresponding inorganic fertilizer variety, the dosage ratio, and the application sequence, generate an initial water and fertilizer plan.

[0107] For example, embodiments of the present invention can integrate all the above analysis results to generate a complete initial water and fertilizer plan for each candidate organic fertilizer variety. Key technical parameters include the selection and application rate of organic fertilizer varieties, the matching inorganic fertilizer varieties and application rates, the organic-inorganic fertilizer ratio, and the timing and method of basal fertilizer application. The resulting initial water and fertilizer plan achieves nutrient supply and demand balance and embodies the scientific fertilization concept of combining organic and inorganic fertilizers and integrating fast-acting and slow-acting fertilizers, providing a high-quality basis for subsequent optimization decisions.

[0108] As another possible implementation, embodiments of the present invention can, based on the target nutrient supply of organic fertilizer and the nutrient content, carbon-nitrogen ratio, decomposition degree, and mineralization rate characteristics of various organic fertilizers, select multiple candidate organic fertilizer varieties from available organic fertilizer resources; based on the target nutrient supply at the basal fertilizer stage and the principle of organic-inorganic fertilizer application, determine the amount of organic fertilizer used at 30-70% of the total nitrogen; based on multiple candidate organic fertilizer varieties and facility meteorological data, combined with the nutrient mineralization characteristics of organic fertilizer, calculate the effective nutrient release amount of each candidate organic fertilizer variety during the vegetable crop production period; compare the target nutrient supply of organic fertilizer with the effective nutrient release amount of each candidate organic fertilizer variety to generate the corresponding inorganic fertilizer supplementation amount for each candidate organic fertilizer variety, and determine the nutrient supply amount at each growth stage; based on each candidate organic fertilizer variety and the corresponding inorganic fertilizer variety, amount, and ratio, generate an initial scheme.

[0109] As another possible implementation, embodiments of the present invention can combine meteorological data and the growth stage of facility crops to determine the initial water supply plan, including: determining the water evapotranspiration of a reference crop under the environmental conditions based on meteorological data; simulating the crop coefficient based on meteorological data and the crop growth stage; multiplying the water evapotranspiration of the reference crop under the environmental conditions and the crop coefficient to obtain the actual crop evapotranspiration, and finally determining the irrigation timing and water volume.

[0110] S106. Based on the initial water and fertilizer plan, crop target data, and decision support system, perform multi-objective rolling decision optimization to generate the final water and fertilizer plan.

[0111] In some embodiments, the present invention can input an initial water and fertilizer plan into a decision support system with a built-in multi-objective optimization algorithm. The system combines the different emphasis requirements of crops on yield, quality, and efficiency, performs rolling optimization calculations, and generates a final water and fertilizer management plan that achieves the best balance among multiple objectives and can be dynamically adjusted according to environmental and crop growth feedback.

[0112] As one possible implementation, step S106 can be specifically implemented as steps S1061-S1068.

[0113] S1061. Based on crop target data, determine the priority weights of optimization targets.

[0114] For example, embodiments of the present invention can establish a priority weighting system for multi-objective optimization based on the yield and quality requirements explicitly stated in crop target data. In specific implementation, the system automatically configures weight parameters according to the user-selected objective: when the yield-priority mode is selected, the system assigns a high weight of 0.7-0.9 to yield-related indicators and a lower weight of 0.1-0.3 to quality indicators; when the quality-priority mode is selected, the system assigns a weight of 0.6-0.8 to quality indicators, and the yield indicator weight is correspondingly reduced to 0.2-0.3. This weighting configuration process ensures that subsequent optimization directions remain consistent with the user's production objectives.

[0115] S1062. Based on facility meteorological data, establish a set of water and fertilizer application constraints, which includes the maximum single irrigation volume limit, nutrient concentration safety threshold, and fertilization frequency constraints.

[0116] For example, embodiments of the present invention can construct a set of constraints for water and fertilizer application based on real-time monitored facility meteorological data and crop growth status. Specifically, these include: a maximum single irrigation volume limit: the maximum single water supply volume determined based on soil permeability, crop root depth, and facility drainage capacity; a nutrient concentration safety threshold: the upper limit of soil solution nutrient concentration to ensure crop root safety and avoid salt damage or nutrient burn; and fertilization frequency constraints: the minimum fertilization interval set based on crop nutrient absorption patterns and agronomic requirements.

[0117] These constraints together constitute the feasible boundary range of the water and fertilizer scheme, ensuring that the recommended scheme is practically operable.

[0118] S1063. Based on the initial water and fertilizer scheme, priority weights, and water and fertilizer application constraint set, multi-objective optimization is performed in the decision support system to obtain multiple candidate water and fertilizer schemes.

[0119] For example, embodiments of the present invention can input an initial water and fertilizer scheme, priority weights, and a set of constraints into the multi-objective optimization engine of a decision support system. This engine employs advanced optimization algorithms to generate 3-5 candidate water and fertilizer schemes with different emphases among objectives such as yield, quality, and cost, while satisfying all constraints. Each candidate scheme performs excellently on a specific objective while achieving acceptable levels on other objectives.

[0120] S1064. Based on multiple candidate water and fertilizer schemes and the digital twin model in the decision support system, conduct pre-simulations for each growth stage to determine the fertilization effect of each candidate water and fertilizer scheme; the fertilization effect includes the effects in three dimensions: yield formation, quality indicators and water and fertilizer utilization efficiency.

[0121] For example, embodiments of the present invention can utilize a digital twin model in a decision support system to virtually simulate the execution of each candidate water and fertilizer scheme at each growth stage. This digital twin model is constructed by coupling a calibrated crop growth model with an environmental control model, and can simulate the complete growth process of crops from transplanting to harvest under different water and fertilizer schemes, and predict the specific performance of each scheme in three dimensions: yield formation, quality indicators, and water and fertilizer use efficiency.

[0122] S1065. Based on the fertilization effects of each candidate water and fertilizer scheme, calculate the evaluation index of each candidate water and fertilizer scheme.

[0123] In some embodiments, the evaluation indicators include the input-output ratio, fertilizer cost, labor cost, fruit sugar content improvement rate, vitamin content improvement rate, marketable fruit rate improvement, water use efficiency, fertilizer partial productivity, and environmental emissions.

[0124] For example, embodiments of the present invention can, based on the results data output by the digital twin pre-simulation, calculate nine specific evaluation indicators for each candidate water and fertilizer scheme under three criterion layers: Economic benefit layer: calculating the output-input ratio (ratio of output value to input cost), direct fertilizer cost, and labor cost for water and fertilizer management; Quality improvement layer: quantifying the increase in fruit sugar content, vitamin content increase rate, and marketable fruit rate; Resource consumption layer: assessing water use efficiency (dry matter yield per unit of water consumed), fertilizer partial productivity (crop yield per unit of nutrient input), and environmental emission intensity (mainly referring to nitrogen and phosphorus leaching risk). These indicators collectively constitute the quantitative basis for judging the merits of the schemes.

[0125] S1066. Based on the evaluation indicators of each candidate water and fertilizer scheme, the fuzzy hierarchical analysis method is used to screen and obtain the optimal scheme.

[0126] For example, embodiments of the present invention can employ fuzzy hierarchical analysis, a multi-attribute decision-making tool, to comprehensively evaluate various candidate solutions. Specifically, a hierarchical structure model is established, a judgment matrix is ​​constructed, the weights of each indicator are calculated, and the uncertainty and contradictions between indicators are handled through fuzzy comprehensive evaluation. Finally, the comprehensive scores of each solution are obtained and ranked, and the solution with the highest score is selected as the optimal water and fertilizer solution.

[0127] S1067. Based on the optimal solution and combined with the growth stage characteristics of the crop to be fertilized, determine the fertilization time, fertilization frequency, irrigation timing and water volume of organic fertilizer and inorganic fertilizer.

[0128] For example, embodiments of the present invention can formulate a detailed implementation plan based on the selected optimal water and fertilizer scheme and in combination with the specific growth stage characteristics of the crop to be fertilized. This includes: determining the specific application time of basal fertilizer and each topdressing; planning the number of fertilizations and the fertilizer ratio for each growth stage; arranging the timing and amount of irrigation each time, forming an implementation schedule with clear time nodes.

[0129] S1068. Based on the optimal solution, as well as the application time, number of applications, irrigation timing and water volume of organic and inorganic fertilizers, generate the final water and fertilizer solution.

[0130] For example, embodiments of the present invention can integrate the specific parameters and execution plan of the optimal water and fertilizer scheme to generate a final water and fertilizer scheme containing complete elements such as fertilizer type, dosage, ratio, application time, and irrigation regime. This scheme not only reflects the scientific decision-making results of multi-objective optimization, but also has practicality and guidance for on-site implementation, providing a direct basis for the precision water and fertilizer management of greenhouse vegetables.

[0131] This invention provides a smart water and fertilizer management method driven by the needs of greenhouse vegetables. Based on greenhouse meteorological data and crop target data, it drives a growth model to dynamically predict the total nutrient demand at each growth stage, effectively overcoming the lag of traditional static decision-making and matching water and fertilizer supply with the actual growth needs of crops. By introducing soil fertility level analysis to determine nutrient recommendation coefficients, it achieves precise verification based on the actual soil fertility capacity, significantly improving the rationality of nutrient input. Furthermore, by combining the principles of organic and inorganic fertilizer application with the mineralization characteristics of organic fertilizer to formulate an initial plan, it ensures the scientific and sustainable nature of nutrient supply. Finally, through a decision support system, multi-objective optimization is used to generate an execution plan, forming a complete decision-making closed loop from prediction and planning to optimization. This solves the problems of insufficient responsiveness and poor environmental adaptability in current greenhouse vegetable water and fertilizer management technologies, improving the accuracy and reliability of water and fertilizer management for greenhouse vegetables.

[0132] Optionally, the intelligent water and fertilizer management method based on demand-driven facility vegetables provided in this embodiment of the invention further includes steps S201-S203 after step S106.

[0133] S201: Real-time multidimensional growth data of the root-plant-canopy parts of the crop to be fertilized are collected through soil profile moisture sensors, crop stem micro-change sensors, and canopy multispectral imagers.

[0134] In some embodiments, the present invention can utilize a multi-type sensor network deployed in the crop growth environment to systematically collect multi-dimensional growth data of various parts of the root-plant-canopy system in real time. Specific implementations include: soil profile moisture sensors: monitoring the volumetric water content of soil at different depths to obtain the dynamic distribution of water in the root zone; crop stem micro-change sensors: continuously measuring the daily variation of stem diameter using high-precision displacement sensing technology, an indicator that directly reflects the crop's internal water status; and a canopy multispectral imager: periodically acquiring multispectral images of the crop canopy and retrieving canopy physiological parameters such as leaf area index and chlorophyll content by analyzing the reflectance characteristics of different wavelength bands. These sensors constitute a complete crop growth status sensing system, providing a data foundation for subsequent analysis.

[0135] S202. Based on multidimensional growth data and the expected data in the digital twin model of the crop to be fertilized, key indicators are compared to determine the comparison results of key indicators.

[0136] In some embodiments, the present invention can input real-time collected multidimensional growth data into a digital twin model of the crop to be fertilized, and accurately compare it with the expected growth data predicted by the model based on current environmental conditions and the implemented water and fertilizer program. Specifically, the system focuses on the matching degree of several key indicators: including the development dynamics of the canopy leaf area index, the daily cycle pattern of stem diameter changes, and the fluctuation pattern of soil moisture in the root zone. By establishing a difference analysis algorithm between actual observed values ​​and model expected values, the degree of deviation between the current crop growth state and the ideal state is quantified.

[0137] S203. If the deviation of a key indicator exceeds the threshold in the comparison results of key indicators, an early warning will be triggered, and the actual growth rate and nutrient absorption efficiency of the crop to be fertilized will be used as reward signals to adjust the amount and frequency of water and fertilizer application in real time.

[0138] In some embodiments, the present invention can immediately trigger an early warning mechanism when the system detects through comparative analysis that the deviation of one or more key growth indicators exceeds a preset threshold. Specific implementation includes: setting deviation thresholds for various indicators, such as a canopy leaf area index deviation exceeding 15%, abnormal daily stem diameter variation reaching 20%, and root zone soil moisture deviating from the target range by 10%; after the early warning is triggered, the system automatically uses the actual growth rate and nutrient absorption efficiency of the crop as reward signals for the reinforcement learning algorithm; based on the feedback of the reward signals, the system adjusts the amount and frequency of water and fertilizer application in real time for subsequent stages, such as appropriately increasing the number of irrigations, adjusting nutrient concentrations, or changing the timing of fertilization.

[0139] Thus, this embodiment of the invention achieves a fundamental shift from static decision-making to dynamic optimization by establishing a closed-loop control mechanism of "monitoring-comparison-early warning-adjustment". The system can perceive the crop growth status in real time, and automatically trigger regulation when the monitoring data deviates from the model prediction, significantly improving the accuracy and anti-interference ability of water and fertilizer management, effectively responding to sudden environmental stresses, and ensuring that crops receive the best water and fertilizer supply throughout their growth period.

[0140] Optionally, the intelligent water and fertilizer management method based on demand-driven facility vegetables provided in this embodiment of the invention further includes steps S301-S306 after step S106.

[0141] S301. The transpiration rate of the crop to be fertilized is monitored in real time by a stem flow meter, the canopy temperature of the crop to be fertilized is monitored in real time by an infrared temperature sensor, and the stomatal conductance change rate is monitored in real time by a stomatal meter.

[0142] For example, embodiments of the present invention can continuously collect key physiological parameters reflecting the crop's water and nutrient status using precision sensing devices deployed in the crop canopy. These parameters include: crop transpiration rate: directly measuring the rate of water transport in the crop stem using a stem flow meter, which directly reflects the intensity of the crop's water metabolism; canopy temperature: non-contact measurement of the average temperature of the crop canopy surface using an infrared temperature sensor; and stomatal conductance change rate: periodically monitoring the trend of changes in leaf stomatal opening using a portable stomameter, which characterizes the activity level of gas exchange in crop leaves.

[0143] S302. If the crop transpiration rate is lower than the preset threshold, irrigation adjustment is triggered.

[0144] For example, embodiments of the present invention can establish a water stress diagnosis mechanism based on transpiration rate. When the crop transpiration rate is detected to be below 20% of the normal value for two consecutive hours, the system determines that the crop is under water stress and automatically triggers the irrigation system adjustment program. Adjustment measures include increasing the irrigation frequency or the amount of irrigation per application to ensure that the crop water supply recovers to an appropriate level.

[0145] S303. Calculate the leaf-air temperature difference based on the canopy temperature; if the leaf-air temperature difference exceeds the set temperature difference, initiate further optimization of the water-fertilizer ratio.

[0146] For example, embodiments of the present invention can calculate the leaf-air temperature difference based on real-time monitored canopy temperature and air temperature data. When the temperature difference exceeds a set threshold of ±2°C and persists for more than 1 hour, the system determines that the crop is under heat stress or has nutrient absorption impairment, and automatically initiates a water-fertilizer ratio re-optimization program. By adjusting the water-nutrient ratio in the water and fertilizer supply, the system improves the crop's water status and nutrient absorption efficiency.

[0147] S304. If the stomatal conductance change rate is abnormal, adjust the nutrient supply intensity.

[0148] For example, embodiments of the present invention can establish a correlation model between the rate of change in stomatal conductance and nutrient absorption efficiency. When stomatal conductance is detected to be below 0.1 mol / m²·s or above 0.4 mol / m²·s and remains abnormal, the system determines that the crop's nutrient absorption mechanism is impaired and automatically adjusts the nutrient supply intensity. For cases of excessively low stomatal conductance, the nutrient concentration is appropriately reduced to avoid root stress; for cases of excessively high stomatal conductance, the nutrient supply is moderately increased to meet the needs of vigorous crop growth.

[0149] S305. Based on crop transpiration rate, leaf-air temperature difference, and stomatal conductance change rate, a weighted fusion is performed to generate quantitative diagnostic results for the crop to be fertilized.

[0150] In some embodiments, the quantitative diagnostic results include thirst or hunger, as well as water or nutrient requirements.

[0151] For example, embodiments of the present invention can construct a multi-index weighted fusion diagnostic model based on the entropy weight method, which systematically integrates three key physiological indicators: crop transpiration rate, leaf-air temperature difference, and stomatal conductance change rate. By calculating the objective weights of each indicator, a quantitative crop status diagnostic result is generated, accurately identifying whether the crop is thirsty, hungry, or in a normal state, and precisely calculating the corresponding water or nutrient requirement adjustment coefficients.

[0152] S306. Based on quantitative diagnostic results, the water and fertilizer supply parameters are finely adjusted in real time through fuzzy control algorithms to achieve precise supply according to the actual physiological needs of crops.

[0153] For example, embodiments of the present invention can design a specialized fuzzy controller based on quantitative diagnostic results to fine-tune water and fertilizer supply parameters in real time. This controller takes crop state diagnostic results as input and, through a preset fuzzy inference rule base, outputs the adjustment amount of water and fertilizer supply parameters, achieving precise supply based on the actual physiological needs of the crop. This control method can effectively handle the uncertainties and nonlinear characteristics in the crop growth process, ensuring that water and fertilizer supply always maintains an optimal match with the actual needs of the crop.

[0154] In some embodiments, the fuzzy controller is a Mamdani-type fuzzy controller with dual inputs of nutrient demand and environmental stress, which couples the nutrient demand prediction results and crop growth status monitoring data to realize real-time dynamic fine-tuning of water and fertilizer supply parameters for facility vegetables. This solves the problem that traditional static fertilization schemes cannot respond to real-time crop growth status and environmental stress. The controller can be embedded into the water and fertilizer integration system of this application to realize closed-loop control from prediction results to execution instructions.

[0155] In some embodiments, the fuzzy controller is a classic two-input dual-output fuzzy control architecture, including five modules: an input module, a fuzzification module, a fuzzy inference rule base, a defuzzification module, and an output module. It adopts the core process of "input fuzzification → fuzzy inference → defuzzification" with a control cycle of 30 minutes. It is coupled in real time with the nutrient demand prediction and water stress coefficient calculation of this application. The inputs are the crop nutrient stress degree and water stress degree, and the outputs are the water supply adjustment amount and nutrient supply adjustment amount, which directly drive the water and fertilizer integrated equipment to perform variable operations.

[0156] For example, the input module collects crop growth status monitoring data and quantifies it into two input variables, both ranging from [0,1] (0 for no stress, 1 for severe stress): Nutrient stress level X1: determined by the deviation between the predicted nutrient requirement and the actual nutrient uptake by the crop. ; where N pred To predict nutrient requirements, N act Measured by a canopy multispectral imager.

[0157] Water stress level X2: Following the logic of water stress coefficient calculation, it is determined by soil matrix potential / effective water content in the root zone and is synchronized in real time with the water stress correction module.

[0158] For example, the fuzzification module uses a triangular membership function to convert the precise values ​​of two input variables into fuzzy sets, each divided into 5 fuzzy levels: {No Stress (NS), Slight Stress (LS), Moderate Stress (MS), Severe Stress (SS), and Extreme Stress (ES)}, ensuring the accuracy and interpretability of the fuzzification process.

[0159] Taking nutrient stress level X1 as an example, the membership function parameters are shown in Table 1. (The parameters for water stress level X2 are the same).

[0160] Table 1 Input Variable Membership Function Parameters

[0161] For example, the fuzzy inference rule base is the core of the controller. It is built based on the nutrient demand prediction results, water stress coefficient and agronomic experience of water and fertilizer management of greenhouse vegetables. It adopts the rule form of IF X1 IS A AND X2 IS B, THEN Y1 IS CAND Y2 IS D, and a total of 25 core rules (5×5) are constructed to cover all stress combinations.

[0162] The output variables are water supply adjustment Y1 and nutrient supply adjustment Y2, both of which take values ​​in the range of [-0.5, 0.5] (-0.5 means a significant decrease, 0 means no change, and 0.5 means a significant increase). The core reasoning rules are shown in Table 2 (the complete rule base is generated by gradient interpolation).

[0163] Table 2 Core Inference Rules for Fuzzy Controllers

[0164] For example, the defuzzification module uses the centroid method to transform the fuzzy set obtained from fuzzy inference into precise adjustment values. This method has high calculation accuracy and meets the actual needs of water and fertilizer adjustment in greenhouse vegetables. The core formula is: ; Among them, y To precisely adjust the amount, yk represents the discrete points of the output variable (11 points at equal intervals in the range [-0.5, 0.5]), and ωij represents the rule trigger strength. To output the membership degree of the fuzzy set.

[0165] For example, the output module converts the precise adjustment amount obtained from defuzzification into control instructions that can be executed by the integrated water and fertilizer system, and couples them with the integrated water and fertilizer control instructions.

[0166] Water supply adjustment: the adjustment ratio of irrigation amount / irrigation frequency, such as Y1=0.2 indicating an increase of 20% in irrigation amount; Nutrient supply adjustment: the adjustment ratio of fertilizer amount / fertilizer concentration, such as Y2=0.3 indicating an increase of 30% in fertilizer concentration; at the same time, the output command can be coupled with the organ-level fertilization scheme of this application to realize the directional fine adjustment of foliar fertilization / root fertilization.

[0167] For example, the fuzzy controller is programmed in Python / C++ and embedded in the microprocessor of the crop nutrient demand prediction system. It communicates in real time with the data acquisition block and the integrated water and fertilizer system. The control cycle is 30 minutes, and it can realize the automated process of data acquisition → stress quantification → fuzzy inference → instruction output.

[0168] Field verification was conducted in a greenhouse tomato experimental field, with a fuzzy control group and a conventional fertilization group as controls. The results showed that the fuzzy control group improved water and fertilizer utilization efficiency by 25%-30%, fruit marketability by 15%-20%, and nitrogen, phosphorus, and potassium nutrient absorption efficiency by 20%-25%, verifying the accuracy and practicality of the controller.

[0169] Thus, this embodiment of the invention achieves a leap from experience-based fertilization to physiological demand-driven fertilization by establishing a precise diagnostic mechanism based on the fusion of multiple physiological indicators. The system can sense the thirst and hunger status of crops in real time, and dynamically fine-tune water and fertilizer parameters through fuzzy control algorithms, significantly improving the accuracy and timeliness of water and fertilizer supply, ensuring that crops can obtain the most suitable water and fertilizer ratio under different growth environments, and effectively improving water and fertilizer utilization efficiency.

[0170] Optionally, the intelligent water and fertilizer management method based on demand-driven facility vegetables provided in this embodiment of the invention further includes steps S401-S404 after step S106.

[0171] S401. Record the diurnal cycle of photosynthesis and nutrient transport in the crop to be fertilized.

[0172] For example, embodiments of the present invention can establish a database of the diurnal cycle patterns of crop photosynthesis and nutrient transport through continuous monitoring and data recording. Specifically, the recorded content includes: the variation patterns of crop photosynthetic rate under daytime light conditions, and the dynamic characteristics of the transport and distribution of assimilated products to various organs at night. Based on in-depth analysis of these physiological processes, the system establishes a scientific allocation strategy that prioritizes water supply during the day and nutrient supply at night. This strategy fully considers the priority of crop needs for different resources at different times.

[0173] S402. Based on the diurnal cycle, a distribution strategy is determined that water supply is the main factor during the day and nutrient supply is the main factor at night.

[0174] S403. Based on the allocation strategy, differentiated parameter settings are made to determine the water and fertilizer supply parameters for daytime and nighttime periods.

[0175] In some embodiments, the water and fertilizer supply parameters during the daytime are mainly high-frequency, low-volume irrigation, including setting the water supply frequency, the amount of water supplied at one time, and the soil moisture target; the water and fertilizer supply parameters during the nighttime are mainly low-frequency, high-volume fertilization, including setting the fertilizer supply frequency and the amount of fertilizer supplied at one time.

[0176] For example, in embodiments of the present invention, the system can configure differentiated water and fertilizer supply parameters for daytime and nighttime periods according to a determined allocation strategy. During the daytime period (usually 6:00-18:00), a high-frequency, low-volume irrigation mode is adopted, with the water supply frequency set at 2-4 times per hour, and the single water supply volume controlled at a suitable level to maintain soil moisture within 70%-80% of field capacity. During the nighttime period (18:00-6:00 the next day), a low-frequency, high-volume fertilization mode is adopted, with the fertilizer supply frequency set at once every 2-3 hours, and the single fertilizer supply volume controlled at 15%-20% of the total demand, ensuring effective accumulation of nutrients in the root zone and preparing nutritional reserves for the crop's growth activities the next day.

[0177] S404. Based on real-time detection of light intensity and temperature, the water and fertilizer supply parameters for daytime and nighttime periods are adjusted in real time to obtain an optimized water and fertilizer scheme and maximize water and fertilizer absorption efficiency.

[0178] For example, embodiments of the present invention can establish an environmentally responsive parameter adjustment mechanism based on real-time monitored light intensity and temperature data. Specifically, when high-temperature, high-light conditions are detected (light intensity greater than 800 μmol / m²·s and temperature above 28°C), the daytime water supply pulse frequency is automatically increased to 4-6 times per hour to alleviate transpiration stress on crops. When the temperature is within the suitable range of 18-25°C, the nighttime fertilizer pulse dosage is increased to 20%-25% of the total requirement, fully utilizing suitable nighttime temperature conditions to promote nutrient absorption and transport.

[0179] For example, embodiments of the present invention can establish a pulse regulation effect evaluation system based on crop growth performance. The system verifies the actual effect of the pulse scheme by monitoring three key indicators: morning leaf expansion rate, midday wilting degree, and nighttime nutrient absorption efficiency. When the morning leaf expansion rate is below 85%, the system automatically increases the baseline amount of daytime water supply pulses; when the midday wilting index exceeds 0.3, the pulse frequency during high-temperature periods is increased; and when the nighttime nutrient absorption efficiency is below 75%, the nutrient ratio of the nighttime pulses is adjusted. Through this continuous effect evaluation and parameter optimization, the pulse regulation scheme is ensured to maintain optimal performance at all times.

[0180] For example, embodiments of the present invention can integrate the results of environmental response adjustment and effect evaluation optimization to generate a final optimized water and fertilizer program. This program precisely specifies water and fertilizer supply parameters for different time periods, including key operational parameters such as pulse frequency, single dose, and nutrient ratio, and is implemented precisely through an intelligent execution system. This differentiated water and fertilizer supply strategy based on the crop's diurnal physiological rhythm effectively aligns with the crop's inherent biological clock, significantly improving water and fertilizer absorption and utilization efficiency, and providing a reliable technical guarantee for achieving high-yield and high-quality crops.

[0181] Thus, this invention establishes a water and fertilizer pulse regulation mechanism based on the diurnal physiological rhythm of crops, achieving precise alignment between water and fertilizer supply and the crop's biological clock. High-frequency irrigation during the day ensures the needs of photosynthesis, while concentrated fertilization at night promotes nutrient absorption. Combined with environmentally responsive dynamic adjustments, this significantly improves water and fertilizer utilization efficiency, effectively solving the resource waste problem caused by traditional uniform irrigation and fertilization, and providing a more precise management and control solution that conforms to the physiological characteristics of crops for greenhouse vegetable production.

[0182] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0183] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0184] Figure 2 This diagram illustrates the structure of a smart water and fertilizer management device based on the demand-driven nature of greenhouse vegetables, according to an embodiment of the present invention. The management device 500 includes a communication module 501 and a processing module 502.

[0185] The communication module 501 is used to acquire facility meteorological data, soil data and crop target data of the crop to be fertilized.

[0186] The processing module 502 is used to predict the total nutrient requirements of the crop at each growth stage based on facility meteorological data, crop target data, and a preset facility vegetable growth model; analyze soil fertility level based on soil data and determine nutrient recommendation coefficients; determine the total nutrient input based on the total nutrient requirements and nutrient recommendation coefficients; determine the initial water and fertilizer scheme for organic and inorganic fertilizers based on the total nutrient input, the principle of organic and inorganic fertilizer application, and the mineralization characteristics of organic fertilizer nutrients; and perform multi-objective rolling decision optimization based on the initial water and fertilizer scheme, crop target data, and the decision support system to generate the final water and fertilizer scheme.

[0187] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. The electronic device 600 includes: a processor 601, a memory 602, and a computer program 603 stored in the memory 602 and executable on the processor 601. When the processor 601 executes the computer program 603, it implements the steps in the above-described method embodiments. Alternatively, when the processor 601 executes the computer program 603, it implements the functions of each module / unit in the above-described device embodiments.

[0188] For example, the computer program 603 may be divided into one or more modules / units, which are stored in the memory 602 and executed by the processor 601 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program 603 in the electronic device 600.

[0189] The processor 601 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0190] The memory 602 can be an internal storage unit of the electronic device 600, such as a hard disk or memory of the electronic device 600. The memory 602 can also be an external storage device of the electronic device 600, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD) card, flash card, etc., equipped on the electronic device 600. Furthermore, the memory 602 can include both internal and external storage units of the electronic device 600. The memory 602 is used to store the computer program and other programs and data required by the terminal. The memory 602 can also be used to temporarily store data that has been output or will be output.

[0191] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. 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. Such 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, and should all be included within the protection scope of the present invention.

Claims

1. A smart water and fertilizer management method based on demand-driven principles for greenhouse vegetables, characterized in that, include: Acquire facility meteorological data, soil data, and crop target data for crops to be fertilized; Based on the facility meteorological data and the crop target data, as well as the preset facility vegetable growth model, the total nutrient requirements of the crop to be fertilized at each growth stage are predicted. Based on the soil data, the soil fertility level was analyzed, and the nutrient recommendation coefficient was determined. The total nutrient input is determined based on the total nutrient requirement and the nutrient recommendation coefficient. Based on the total amount of nutrients input and the principle of organic and inorganic fertilizer application, combined with the mineralization characteristics of organic fertilizer nutrients, the initial water and fertilizer schemes for organic and inorganic fertilizers are determined. Based on the initial water and fertilizer plan, the crop target data, and the decision support system, multi-objective rolling decision optimization is performed to generate the final water and fertilizer plan.

2. The intelligent water and fertilizer management method based on demand-driven greenhouse vegetable production according to claim 1, characterized in that, The facility meteorological data includes light intensity, air temperature and air humidity; the soil data includes soil fertility level, initial soil nutrient content and soil moisture data; and the crop target data includes target yield and / or target quality. The method of predicting the total nutrient requirements of the crop at each growth stage based on the facility meteorological data, the crop target data, and a preset facility vegetable growth model includes: The meteorological data of the facility is used as the core driving variable and input into the facility vegetable growth model to simulate the photosynthetic process and dry matter accumulation process of the crop to be fertilized from transplanting to harvest. During the simulation, based on the target crop data, the biomass distribution ratio of each organ of the crop to be fertilized is determined, and the total amount of dry matter required by the crop to be fertilized is calculated. Based on the nutrient absorption patterns of the crops to be fertilized, and combined with the dry matter accumulation process at each growth stage, the absorption amount and intensity of nitrogen, phosphorus, and potassium nutrients at each growth stage are calculated. Based on the total dry matter required by the crop to be fertilized, as well as the absorption amount and intensity of nitrogen, phosphorus, and potassium nutrients at each growth stage, the total nutrient requirement is calculated, which includes the total nitrogen requirement, the total phosphorus requirement, and the total potassium requirement.

3. The intelligent water and fertilizer management method based on demand-driven greenhouse vegetable production according to claim 1, characterized in that, The soil data includes organic matter content, available nitrogen content, available phosphorus content, available potassium content, pH value, soil bulk density, and soil texture. The process of analyzing soil fertility levels and determining nutrient recommendation coefficients based on the soil data includes: Based on the soil data, the soil fertility level is determined using a comprehensive soil fertility evaluation method; the soil fertility level includes high fertility, medium fertility, and low fertility. Nutrient recommendation coefficients are determined based on soil fertility levels and a pre-established mapping relationship between soil fertility levels and nutrient recommendation coefficients.

4. The intelligent water and fertilizer management method based on demand-driven greenhouse vegetable production according to claim 1, characterized in that, The determination of the total nutrient input based on the total nutrient requirement and the nutrient recommendation coefficient includes: Based on the total nutrient requirement and the dynamic nutrient allocation principle based on the crop growth stage, the time period is divided according to the physiological requirements of different growth stages in each growth period, and multiple growth stages and the nutrient requirement allocation of each growth stage are determined. Based on soil fertility levels, the nutrient recommendation coefficients are modified to be specific to the growth stage, generating dynamic recommendation coefficients for each growth stage. The nutrient demand allocation and dynamic recommendation coefficient of each reproductive stage are weighted and integrated to calculate the dynamic nutrient input of each reproductive stage. Based on the facility meteorological data, environmental factors are generated; these environmental factors are used to characterize the degree of influence of the facility meteorological data on the total nutrient input. Based on the aforementioned environmental factors, the dynamic nutrient input at each growth stage is adjusted to obtain the total nutrient input.

5. The intelligent water and fertilizer management method based on demand-driven greenhouse vegetable production according to claim 1, characterized in that, The initial water and fertilizer scheme for organic and inorganic fertilizers, determined based on the total nutrient input, the principle of organic-inorganic fertilizer application, and the mineralization characteristics of organic fertilizer nutrients, includes: Based on the total amount of nutrients input, the nutrients are allocated according to the preset ratio of base fertilizer to topdressing to determine the target nutrient supply for the base fertilizer stage. Based on the target nutrient supply amount for the basal fertilizer stage and the principle of organic and inorganic fertilizer application, the proportion of nutrients that organic fertilizer should provide in the basal fertilizer is analyzed, and the target fertilizer supply amount of organic fertilizer is determined. Based on the target fertilizer supply of organic fertilizer, as well as the nutrient content, carbon-nitrogen ratio, decomposition degree and mineralization rate characteristics of various organic fertilizers, multiple candidate organic fertilizer varieties were screened from available organic fertilizer resources. Based on multiple candidate organic fertilizer varieties and the meteorological data of the facility, combined with the mineralization characteristics of organic fertilizer nutrients, the effective nutrient release of each candidate organic fertilizer variety during the basal fertilizer period is calculated. The target fertilizer supply amount of the organic fertilizer and the effective nutrient release amount of each candidate organic fertilizer variety are compared to generate the corresponding inorganic fertilizer supplement amount for each candidate organic fertilizer variety. Based on the inorganic fertilizer supplementation amount corresponding to each candidate organic fertilizer variety and the mineralization rate characteristics of each candidate organic fertilizer variety, inorganic fertilizer is screened to obtain inorganic fertilizer varieties that match each candidate organic fertilizer variety. Based on each candidate organic fertilizer variety, the corresponding inorganic fertilizer variety, the dosage ratio, and the application sequence, the initial water and fertilizer scheme is generated.

6. The intelligent water and fertilizer management method based on demand-driven greenhouse vegetable production according to claim 1, characterized in that, The step of performing multi-objective decision optimization based on the initial water and fertilizer plan, the crop target data, and the decision support system to generate the final water and fertilizer plan includes: Based on the crop target data, the priority weights of the optimization targets are determined; Based on facility meteorological data, a set of water and fertilizer application constraints is established, which includes a maximum single irrigation volume limit, a nutrient concentration safety threshold, and a fertilization frequency constraint. Based on the initial water and fertilizer scheme, the priority weights, and the set of water and fertilizer application constraints, multi-objective optimization is performed in the decision support system to obtain multiple candidate water and fertilizer schemes. Based on the multiple candidate water and fertilizer schemes and the digital twin model in the decision support system, simulations are performed for each growth stage to determine the fertilization effect of each candidate water and fertilizer scheme; the fertilization effect includes the effects in three dimensions: yield formation, quality indicators, and water and fertilizer use efficiency. Based on the fertilization effects of each candidate water and fertilizer scheme, the evaluation indicators of each candidate water and fertilizer scheme are calculated. The evaluation indicators include the output-input ratio, fertilizer cost, labor cost, fruit sugar content improvement rate, vitamin content improvement rate, marketable fruit rate improvement, water use efficiency, fertilizer partial productivity, and environmental emission indicators. Based on the evaluation indicators of each candidate water and fertilizer scheme, the fuzzy hierarchical analysis method is used to screen and obtain the optimal scheme. Based on the optimal scheme, and combined with the growth stage characteristics of the crop to be fertilized, the fertilization time, fertilization frequency, irrigation timing and water volume of organic fertilizer and inorganic fertilizer are determined. Based on the optimal solution, and the application time, frequency, irrigation timing and volume of the organic and inorganic fertilizers, the final water and fertilizer solution is generated.

7. The intelligent water and fertilizer management method based on demand-driven greenhouse vegetable production according to any one of claims 1 to 6, characterized in that, After generating the final water and fertilizer plan by performing multi-objective decision optimization based on the initial water and fertilizer plan, the crop target data, and the decision support system, the process further includes: The soil profile moisture sensor, crop stem micro-change sensor, and canopy multispectral imager are used to collect multidimensional growth data of the root-plant-canopy parts of the crop to be fertilized in real time. Based on the multidimensional growth data and the expected data in the digital twin model of the crop to be fertilized, key indicators are compared to determine the comparison results of the key indicators. If the deviation of a key indicator exceeds the threshold in the comparison results, an early warning is triggered, and the actual growth rate and nutrient absorption efficiency of the crop to be fertilized are used as reward signals to adjust the amount and frequency of water and fertilizer application in real time.

8. The intelligent water and fertilizer management method based on demand-driven greenhouse vegetable production according to any one of claims 1 to 6, characterized in that, After generating the final water and fertilizer plan by performing multi-objective decision optimization based on the initial water and fertilizer plan, the crop target data, and the decision support system, the process further includes: The transpiration rate of the crop to be fertilized is monitored in real time by a stem flow meter, the canopy temperature of the crop to be fertilized is monitored in real time by an infrared temperature sensor, and the stomatal conductance change rate is monitored in real time by a stomatometer. If the crop transpiration rate is lower than a preset threshold, irrigation adjustments will be triggered. Based on the canopy temperature, calculate the leaf-air temperature difference; if the leaf-air temperature difference exceeds the set temperature difference, initiate further optimization of the water-fertilizer ratio. If the rate of change in stomatal conductance is abnormal, adjust the nutrient supply intensity. Based on the crop transpiration rate, leaf-air temperature difference, and stomatal conductance change rate, a weighted fusion is performed to generate a quantitative diagnostic result for the crop to be fertilized. The quantitative diagnostic result includes thirst or hunger, as well as water or nutrient requirements. Based on the quantitative diagnostic results, the water and fertilizer supply parameters are finely adjusted in real time using a fuzzy control algorithm to achieve precise supply according to the actual physiological needs of the crop.

9. The intelligent water and fertilizer management method based on demand-driven greenhouse vegetable production according to any one of claims 1 to 6, characterized in that, After generating the final water and fertilizer plan by performing multi-objective decision optimization based on the initial water and fertilizer plan, the crop target data, and the decision support system, the process further includes: Record the diurnal cycle patterns of photosynthesis and nutrient transport in the crop to be fertilized; Based on the aforementioned diurnal cycle pattern, an allocation strategy is determined that prioritizes daytime water supply and nighttime nutrient supply. Based on the allocation strategy, differentiated parameters are set to determine the water and fertilizer supply parameters for daytime and nighttime periods. The water and fertilizer supply parameters for daytime periods are mainly high-frequency, low-volume irrigation, including setting the water supply frequency, single water supply volume, and soil moisture target. The water and fertilizer supply parameters for nighttime periods are mainly low-frequency, high-volume fertilization, including setting the fertilization frequency and single fertilization volume. Based on real-time monitoring of light intensity and temperature, the water and fertilizer supply parameters for daytime and nighttime periods are adjusted in real time to obtain an optimized water and fertilizer scheme and maximize water and fertilizer utilization efficiency.

10. A decision support system, characterized in that, The decision support system includes an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor is used to call and run the computer program stored in the memory to perform the method as described in any one of claims 1 to 9.

Images