Nutrient solution refined supply control method based on aeroponic muskmelon growth prediction

Abstract

The invention discloses a nutrient solution refined supply control method based on aeroponic muskmelon growth prediction, and relates to the related field of agricultural intellectualization, and the method comprises the steps: carrying out the data collection, and obtaining the growth environment data; carrying out growth analysis to obtain growth characteristic parameters, carrying out nutritional requirement calculation, and determining an atomized nutrient solution coefficient; constructing a growth prediction model, obtaining a growth prediction trend curve, carrying out supply analysis, and formulating a nutrient solution supply strategy; the growth of the muskmelon is monitored in real time, the atomized nutrient solution coefficient is dynamically adjusted according to a growth monitoring log, and a second atomized nutrient solution coefficient is generated; a nutrient solution supply strategy is updated and optimized, and self-adaptive refined supply control is carried out. The technical problem that fine control is insufficient due to the fact that existing mist culture melon nutrient solution supply control lacks real-time performance, accuracy and dynamic adjustment capacity is solved, and the technical effect of fine control is achieved through real-time, accurate and dynamic adjustment of the nutrient solution.

Description

Technical Field

[0001] The present application relates to fields related to agricultural intelligence, and in particular to a method for fine-grained supply control of nutrient solution based on growth prediction of fog-cultured melons. Background Art

[0002] In modern agriculture, greenhouse cultivation technology has been widely used. Among them, aeroponics has become an important soilless cultivation method due to its advantages such as efficient use of water resources and nutrients and promotion of plant growth rate. The control of nutrient solution supply during the growth of aeroponic melons is directly related to the yield, quality and economic benefits of melons. In the traditional aeroponic melon nutrient solution supply control method, a fixed nutrient solution ratio and supply frequency are usually adopted. This method is often based on experience or a fixed schedule, lacks adaptability to melon growth characteristics and environmental changes, cannot reflect the changes in nutrient demand during melon growth in real time, and lacks dynamic adjustment capabilities.

[0003] Among the current related technologies, the supply control of nutrient solution for mist-cultured melon lacks real-time, accuracy and dynamic adjustment capabilities, leading to technical problems such as insufficient refined control. Summary of the invention

[0004] The present application provides a method for fine-grained supply control of nutrient solution based on growth prediction of mist-cultured melons. The method collects environmental data of the melon growing area in real time through a sensor network, analyzes the growth of the melons, calculates nutritional requirements based on the analysis results, and determines the coefficient of the atomized nutrient solution. At the same time, a growth prediction model is constructed, and growth characteristic parameters are synchronized to the model to obtain a growth prediction trend curve. Supply analysis is performed based on the curve, and a nutrient solution supply strategy is formulated. During the execution process, the growth of the melons is monitored in real time, a growth monitoring log is generated, and the coefficient of the atomized nutrient solution is dynamically adjusted according to the log, and other technical means are used to achieve the technical effect of fine-grained control through real-time, accurate and dynamic adjustment of the nutrient solution.

[0005] The present application provides a method for finely controlling the supply of nutrient solution based on the growth prediction of aeroponic melons, comprising:

[0006] Data of the melon growth area is collected through a sensor network to obtain growth environment data; growth analysis of the melon is performed according to the growth environment data to obtain growth characteristic parameters, nutrient demand is calculated according to the growth characteristic parameters, and the atomized nutrient solution coefficient is determined; a growth prediction model is constructed, and the growth characteristic parameters are synchronized to the growth prediction model according to the growth environment data to obtain a growth prediction trend curve, and the atomized nutrient solution coefficient is mapped to the growth prediction trend curve based on the growth characteristic parameters to perform supply analysis and formulate a nutrient solution supply strategy; the nutrient solution supply strategy is executed to monitor the growth of the melon in real time, and a growth monitoring log is generated, and the atomized nutrient solution coefficient is dynamically adjusted according to the growth monitoring log to generate a second atomized nutrient solution coefficient; the nutrient solution supply strategy is updated and optimized based on the second atomized nutrient solution coefficient, and a nutrient solution supply optimization strategy is generated to perform adaptive and refined supply control of the nutrient solution for fog culture melons.

[0007] In a possible implementation, growth analysis is performed on the melon according to the growth environment data to obtain growth characteristic parameters, and the following processing is performed:

[0008] Retrieve the historical growth data record log of the melons in the melon growth area, match the growth environment data in combination with the historical growth data record log, and obtain a full growth fluctuation data set of the melons, wherein the full growth fluctuation data set includes multiple growth stages and multiple growth cycles; perform growth detection on the multiple growth stages according to the multiple growth cycles, generate detection results, and determine the real-time growth stage of the melon according to the detection results; perform growth impact analysis on the melons based on the real-time growth stage to obtain multiple impact factors; perform melon growth evaluation based on the multiple impact factors to obtain a growth impact evaluation result, and perform growth characteristic analysis on the growth impact evaluation result based on the multiple growth cycles to obtain the growth characteristic parameters.

[0009] In a possible implementation, the nutrient requirement is calculated according to the growth characteristic parameters, the atomized nutrient solution coefficient is determined, and the following processing is performed:

[0010] According to the real-time growth stage and the growth characteristic parameters, a demand analysis is performed to determine multiple nutritional demand elements, wherein the multiple nutritional demand elements include a nutrient solution atomization frequency; based on the multiple nutritional demand elements, weights are assigned according to the growth impact assessment results to obtain multiple weight coefficients; nutritional demand calculations are performed according to the nutrient solution atomization frequency based on the multiple weight coefficients combined with the multiple nutritional demand elements to obtain an atomized nutrient solution concentration coefficient; nutritional demand for melons is determined based on the atomized nutrient solution concentration coefficient combined with the real-time growth stage to determine the atomized nutrient solution coefficient.

[0011] In a possible implementation, the nutrient requirement of the melon is determined based on the atomized nutrient solution concentration coefficient in combination with the real-time growth stage, the atomized nutrient solution coefficient is determined, and the following processing is performed:

[0012] Based on the atomized nutrient solution concentration coefficient, simulated nutrient supply is performed to generate a simulated nutrient supply result; based on the simulated nutrient supply result, it is determined whether the real-time growth state of the melon in the real-time growth stage meets the expected growth trend value, and the real-time growth stage includes the real-time growth state; if the real-time growth state of the melon in the real-time growth stage does not meet the expected growth trend value, an adjustment instruction is generated, and the atomization frequency of the nutrient solution is dynamically adjusted through the adjustment instruction to obtain the atomization adjustment frequency of the nutrient solution; the atomized nutrient solution concentration coefficient is updated and calculated according to the atomization adjustment frequency of the nutrient solution to obtain the atomized nutrient solution coefficient; if the real-time growth state of the melon in the real-time growth stage meets the expected growth trend value, a control instruction is generated, and the atomization control of the nutrient solution is performed according to the atomized nutrient solution concentration coefficient through the control instruction to obtain the atomized nutrient solution coefficient.

[0013] In a possible implementation, a growth prediction model is constructed, and according to the growth environment data, the growth characteristic parameters are synchronized to the growth prediction model to obtain a growth prediction trend curve, and the following processing is performed:

[0014] Based on the growth environment data, the growth time series of melons is collected to obtain multiple growth time nodes; the multiple growth time nodes are serialized according to the data collection time series to obtain a growth time series; the growth characteristic parameters are trained and verified according to the growth time series using a long short-term memory network to construct a growth prediction model; multi-dimensional prediction is performed through the growth prediction model to generate multi-dimensional growth prediction data, and the multi-dimensional growth prediction data includes leaf area prediction data, fruit volume prediction data, and plant height prediction data; according to the multiple growth cycles, growth prediction is performed based on the leaf area prediction data to obtain a leaf area prediction trend; according to the multiple growth cycles, growth prediction is performed based on the fruit volume prediction data to obtain a fruit expansion rate prediction trend; according to the multiple growth cycles, growth prediction is performed based on the plant height prediction data to obtain an overall growth rate prediction trend; according to the multiple growth stages, the leaf area prediction trend, the fruit expansion rate prediction trend, and the overall growth rate prediction trend are associated and integrated to obtain the growth prediction trend curve.

[0015] In a possible implementation, based on the growth characteristic parameters, the atomized nutrient solution coefficient is mapped to the growth prediction trend curve to perform supply analysis, formulate a nutrient solution supply strategy, and perform the following processing:

[0016] Based on the growth characteristic parameters, the atomized nutrient solution coefficient is mapped to the leaf area prediction trend for supply analysis to generate first supply and demand data; based on the growth characteristic parameters, the atomized nutrient solution coefficient is mapped to the fruit expansion rate prediction trend for supply analysis to generate second supply and demand data; based on the growth characteristic parameters, the atomized nutrient solution coefficient is mapped to the overall growth rate prediction trend for supply analysis to generate third supply and demand data; based on the first supply and demand data, the second supply and demand data, and the third supply and demand data, synchronous adjustments are made in combination with the multiple growth stages to generate multi-stage supply results, dynamic adjustments are made according to the multi-stage supply results, and the nutrient solution supply strategy is formulated.

[0017] In a possible implementation, the nutrient solution supply strategy is executed to monitor the growth of the melon in real time, a growth monitoring log is generated, the atomized nutrient solution coefficient is dynamically adjusted according to the growth monitoring log, a second atomized nutrient solution coefficient is generated, and the following processing is performed:

[0018] Execute the nutrient solution supply strategy to perform real-time growth environment monitoring and obtain real-time growth environment change data; execute the nutrient solution supply strategy to perform real-time growth characteristic monitoring and obtain real-time growth characteristic change data; determine the monitoring time interval based on the multiple growth cycles, record the growth according to the monitoring time interval in combination with the real-time growth environment change data and the real-time growth characteristic change data, and obtain a growth monitoring log; perform overlay traversal on the growth monitoring log to generate an overlay traversal result, perform anomaly detection according to the overlay traversal result, and generate multiple anomaly labels; map the multiple anomaly labels to the growth monitoring log for matching, determine multiple abnormal growth data, dynamically adjust the atomized nutrient solution coefficient according to the multiple abnormal growth data, and generate the second atomized nutrient solution coefficient.

[0019] In a possible implementation, the multiple abnormal labels are mapped to the growth monitoring log for matching, multiple abnormal growth data are determined, the atomized nutrient solution coefficient is dynamically adjusted according to the multiple abnormal growth data, the second atomized nutrient solution coefficient is generated, and the following processing is performed:

[0020] The multiple abnormal labels are used as indexes to traverse the growth monitoring log for abnormality determination, and it is determined whether the growth monitoring log has values ​​that are not in the preset growth interval; if the growth monitoring log has values ​​that are not in the preset growth interval, the growth monitoring data that are not in the preset growth interval are extracted, and the growth monitoring data that are not in the preset growth interval are matched with the multiple abnormal labels, and multiple abnormal growth data are determined according to the matching results; leaf area analysis is performed based on the multiple abnormal growth data in combination with the growth prediction trend curve to obtain leaf area abnormality data; fruit enlargement analysis is performed based on the multiple abnormal growth data in combination with the growth prediction trend curve to obtain fruit enlargement abnormality data; overall growth analysis is performed based on the multiple abnormal growth data in combination with the growth prediction trend curve to obtain growth rate abnormality data; the leaf area abnormality data, the fruit enlargement abnormality data, and the growth rate abnormality data are integrated, and the atomized nutrient solution coefficient is adjusted and updated according to the integration result to obtain the second atomized nutrient solution coefficient.

[0021] The method for fine-grained supply control of nutrient solution based on growth prediction of mist-cultured melon proposed in this application first collects data on the melon growth area through a sensor network to obtain growth environment data, then analyzes the growth of the melon based on the growth environment data to obtain growth characteristic parameters, calculates nutrient demand according to the growth characteristic parameters, determines the atomized nutrient solution coefficient, then constructs a growth prediction model, synchronizes the growth characteristic parameters to the growth prediction model according to the growth environment data, obtains a growth prediction trend curve, maps the atomized nutrient solution coefficient to the growth prediction trend curve based on the growth characteristic parameters for supply analysis, formulates a nutrient solution supply strategy, and then executes the nutrient solution supply strategy to monitor the growth of the melon in real time, generates a growth monitoring log, dynamically adjusts the atomized nutrient solution coefficient according to the growth monitoring log, generates a second atomized nutrient solution coefficient, and finally updates and optimizes the nutrient solution supply strategy based on the second atomized nutrient solution coefficient, generates a nutrient solution supply optimization strategy, and performs adaptive fine-grained supply control of the nutrient solution for mist-cultured melon, thereby achieving the technical effect of fine control through real-time, accurate and dynamic adjustment of the nutrient solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the method according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0023] Figure 1 A schematic flow chart of a method for fine-grained nutrient solution supply control based on aeroponic melon growth prediction provided in an embodiment of the present application.

[0024] Figure 2 A schematic diagram of a flow chart for determining the coefficient of atomized nutrient solution in a method for finely controlling the supply of nutrient solution based on aeroponic melon growth prediction provided in an embodiment of the present application. DETAILED DESCRIPTION

[0025] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0026] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0027] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0028] The present application embodiment provides a method for finely controlling the supply of nutrient solution based on the growth prediction of aeroponic melons. Figure 1 As shown, the method includes:

[0029] Step S100, collecting data from the melon growing area through a sensor network to obtain growth environment data. Specifically, a sensor network is arranged in the melon growing area, and the sensor network is a network composed of multiple sensors, which is used to monitor and collect environmental data of the melon growing area in real time, including a temperature sensor, a humidity sensor, a light intensity sensor, a CO2 concentration sensor, etc. The sensor network collects environmental data of the melon growing area in real time, such as temperature, humidity, light intensity, CO2 concentration, etc., to obtain the growth environment data.

[0030] Step S200, performing growth analysis on the melon according to the growth environment data, obtaining growth characteristic parameters, calculating the nutritional requirements according to the growth characteristic parameters, and determining the coefficient of the atomized nutrient solution. Specifically, based on the collected growth environment data, using an agricultural data analysis algorithm, the growth characteristic parameters of the melon are extracted. The growth characteristic parameters are parameters that reflect the growth status and trend of the melon, such as growth rate, leaf area index, biomass accumulation, etc. According to the growth characteristic parameters of the melon and the nutritional requirement law of the melon growth, the types and quantities of the nutritional elements required by the melon at different growth stages are calculated. Based on the results of the nutritional requirement calculations and combined with the ratio and supply method of the atomized nutrient solution, the atomized nutrient solution coefficient, that is, the concentration and ratio of each nutritional element in the nutrient solution, is determined.

[0031] In a possible implementation, the growth of the melon is analyzed according to the growth environment data to obtain growth characteristic parameters, and step S200 further includes step S210, calling the historical growth data record log of the melon in the melon growth area, matching the growth environment data with the historical growth data record log, and obtaining the full growth fluctuation data set of the melon, wherein the full growth fluctuation data set includes multiple growth stages and multiple growth cycles. Specifically, the historical growth data record log of the melon in the melon growth area is called from the database, and these data include the growth parameters of the melon (such as plant height, leaf area, fruit size, etc.), growth environment data (such as temperature, humidity, light, etc.) and corresponding growth stages and timestamps in the past period of time. The currently collected growth environment data is matched with the historical growth data record log to find the historical growth conditions similar to the current environment, and based on the matching results, a full growth fluctuation data set including multiple growth stages and multiple growth cycles is constructed. Step S220, traverse the multiple growth stages according to the multiple growth cycles to perform growth detection, generate detection results, and determine the real-time growth stage of the melon according to the detection results. Specifically, multiple growth cycles in the full growth fluctuation data set are traversed in chronological order. In each growth cycle, the current growth stage of the melon (such as seedling stage, vine extension stage, flowering and fruiting stage, fruit expansion stage, etc.) is detected according to the growth parameters and environmental conditions of the melon. The detection results of each growth cycle are recorded to form a detection result set. Step S230, based on the real-time growth stage, growth impact analysis is performed on the melon to obtain multiple influencing factors. Specifically, based on the real-time growth stage and the detection result set, various factors affecting melon growth (such as temperature, humidity, light intensity, nutrient supply, etc.) are identified. Each influencing factor is quantitatively evaluated to determine its specific degree of influence on melon growth, and the quantified influencing factors are recorded to form an influencing factor set. Step S240, melon growth is evaluated according to the multiple influencing factors to obtain a growth impact evaluation result, and growth characteristics analysis is performed on the growth impact evaluation result based on the multiple growth cycles to obtain the growth characteristic parameters. Specifically, based on the influencing factor set and the growth impact evaluation model, the growth of the melon is evaluated. Parameters reflecting the growth characteristics of melons (such as growth rate, biomass accumulation rate, nutrient absorption efficiency, etc.) are extracted from the evaluation results. The extracted growth characteristic parameters are recorded for nutrient demand calculation and nutrient solution supply strategy formulation. This implementation method uses historical data and current environmental data for matching and analysis to accurately predict the growth of melons and improve the accuracy of obtaining growth characteristic parameters.

[0032] In a possible implementation, the nutrient requirement calculation is performed according to the growth characteristic parameters, and the atomized nutrient solution coefficient is determined. Step S200 further includes step S250, and the demand analysis is performed according to the real-time growth stage combined with the growth characteristic parameters to determine multiple nutrient requirement elements, and the multiple nutrient requirement elements include the nutrient solution atomization frequency. Specifically, according to the real-time growth stage of the melon determined in step S220, the physiological characteristics and growth requirements of the melon at this stage are analyzed. The growth characteristic parameters (such as growth rate, biomass accumulation rate, etc.) obtained in step S240 are combined with the real-time growth stage to further refine the nutrient requirements. Based on the above analysis, the main nutrient requirement elements required by the melon at the current growth stage are determined, including macroelements such as nitrogen, phosphorus, and potassium, as well as trace elements such as calcium, magnesium, iron, sulfur, zinc, copper, manganese, molybdenum, and boron (or ion concentrations such as nitrate, ammonium, calcium ion, dihydrogen phosphate, potassium ion, magnesium ion, and sulfate). In addition, the nutrient solution atomization frequency, that is, the number of times the nutrient solution is atomized per unit time, needs to be considered. The nutrient solution atomization frequency affects the melon's absorption efficiency of the nutrient solution. Step S260, weights are assigned based on the multiple nutritional requirement elements according to the growth impact assessment results to obtain multiple weight coefficients. Specifically, according to the growth impact assessment results obtained in step S240, the specific degree of impact of each influencing factor on melon growth is determined. According to the importance of each influencing factor to melon growth and the specific requirements for nutritional elements in the current growth stage, a weight coefficient is assigned to each nutritional requirement element. The weight coefficient reflects the relative contribution of each nutritional requirement element to melon growth in the current growth stage. Step S261:

[0033] S270, calculate the nutrient demand according to the nutrient solution atomization frequency in combination with the multiple weight coefficients, and obtain the atomized nutrient solution concentration coefficient. Specifically, the weight coefficient determined in step S260 is combined with the nutrient demand element determined in step S250 to form a weighted nutrient demand. Considering the effect of the nutrient solution atomization frequency on the absorption efficiency of the melon, the atomization frequency is used as one of the calculation factors. Based on the weighted nutrient demand and the nutrient solution atomization frequency, the atomized nutrient solution concentration coefficient is calculated by a formula. This coefficient reflects the nutrient solution concentration required by the melon at the current growth stage and atomization frequency. The calculation formula for the atomized nutrient solution concentration coefficient is: C nutrient =N demand / (V mist ×F mist ), where C nutrient is the concentration coefficient of the atomized nutrient solution, N demand is the nutritional requirement calculated based on growth characteristic parameters, V mist is the volume of nutrient solution atomized each time, F mistis the frequency of atomization of the nutrient solution. Step S280, judging the nutritional needs of the melon based on the atomized nutrient solution concentration coefficient in combination with the real-time growth stage, and determining the atomized nutrient solution coefficient. Specifically, a nutritional need judgment is made according to the real-time growth stage of the melon in combination with the calculated atomized nutrient solution concentration coefficient. Based on the above judgment, the atomized nutrient solution coefficient required for the melon at the current growth stage and atomization frequency is determined. This coefficient integrates multiple factors such as the growth needs of the melon, nutrient requirement elements, weight coefficients, and nutrient solution atomization frequency. This implementation method analyzes the nutritional needs of melons at different growth stages and atomization frequencies through demand analysis, weight allocation, and nutrient requirement calculation, thereby improving the accuracy of determining the atomized nutrient solution coefficient.

[0034] like Figure 2As shown, in a possible implementation, the nutrient demand of the melon is determined based on the atomized nutrient solution concentration coefficient combined with the real-time growth stage, and the atomized nutrient solution coefficient is determined. Step S280 further includes step S281, simulating nutrient supply based on the atomized nutrient solution concentration coefficient to generate a simulated nutrient supply result. Specifically, all data of the current melon growth environment are collected, including temperature, humidity, light intensity, etc., and the atomized nutrient solution concentration coefficient. Using the growth model in agricultural science, the above data is input to construct a simulated environment for melon growth. In the simulated environment, virtual nutrient supply is performed according to the current atomized nutrient solution concentration coefficient to simulate the growth of melon under specific conditions. The growth data of the melon during the simulation process, including growth rate, biomass accumulation, etc., is recorded to generate a simulated nutrient supply result. Step S282, based on the simulated nutrient supply result, it is judged whether the real-time growth state of the melon in the real-time growth stage meets the expected growth trend value, and the real-time growth stage includes the real-time growth state. Specifically, according to the growth characteristics and historical data of the melon, the expected growth trend value in the current growth stage is set, including key indicators such as growth rate and biomass accumulation. The melon growth data in the simulated nutrient supply result is compared with the expected growth trend value to evaluate the real-time growth status of the melon. If the real-time growth state differs greatly from the expected growth trend value, it means that there may be a problem with the current nutrient supply strategy; if the two are close or consistent, it means that the current nutrient supply strategy is effective. Step S283, if the real-time growth state of the melon in the real-time growth stage does not meet the expected growth trend value, an adjustment instruction is generated, and the nutrient solution atomization frequency is dynamically adjusted through the adjustment instruction to obtain the nutrient solution atomization adjustment frequency. Specifically, when it is judged that the real-time growth state does not meet the expected growth trend value, an adjustment instruction is generated according to the simulated nutrient supply result and growth characteristic analysis. According to the adjustment instruction, the nutrient solution atomization frequency is dynamically adjusted to increase or decrease the supply of nutrient solution, thereby adjusting the growth state of the melon. Step S284, updating and calculating the atomized nutrient solution concentration coefficient according to the nutrient solution atomization adjustment frequency to obtain the atomized nutrient solution coefficient. Specifically, updating the atomized nutrient solution concentration coefficient according to the adjusted nutrient solution atomization frequency. Resimulating nutrient supply using the updated atomized nutrient solution concentration coefficient to verify the validity of the new atomized nutrient solution concentration coefficient. Step S285, if the real-time growth state of the melon in the real-time growth stage meets the expected growth trend value, generating a control instruction, and controlling the atomization of the nutrient solution according to the atomized nutrient solution concentration coefficient through the control instruction to obtain the atomized nutrient solution coefficient. Specifically, generating a control instruction when it is determined that the real-time growth state meets the expected growth trend value. According to the control instruction, using an atomization device, the nutrient solution is atomized into tiny particles, and supplied to the melon according to the nutrient solution atomization adjustment frequency.This implementation method can promptly detect and correct deviations in nutrient supply through simulation of nutrient supply and real-time growth status assessment, thus ensuring the effectiveness of nutrient supply strategies.

[0035] Step S300, construct a growth prediction model, synchronize the growth characteristic parameters to the growth prediction model according to the growth environment data, obtain a growth prediction trend curve, map the atomized nutrient solution coefficient to the growth prediction trend curve based on the growth characteristic parameters to perform supply analysis, and formulate a nutrient solution supply strategy. Specifically, a melon growth prediction model is constructed using a machine learning algorithm combined with historical growth data. The growth characteristic parameters and growth environment data collected in real time are synchronized to the growth prediction model to generate a future growth prediction trend curve for the melon, reflecting the growth status and trend of the melon at different time points. According to the growth prediction trend curve and the atomized nutrient solution coefficient, a nutrient solution supply strategy is formulated, including supply time, supply amount, supply method, etc.

[0036] In a possible implementation, a growth prediction model is constructed, and according to the growth environment data, the growth characteristic parameters are synchronized to the growth prediction model to obtain a growth prediction trend curve. Step S300 further includes step S310, which collects the growth time series of melons based on the growth environment data to obtain multiple growth time nodes. Specifically, the environmental data of the melon growth area, such as temperature, humidity, light intensity, etc., are regularly collected through a sensor network (such as a temperature sensor, a humidity sensor, a light sensor, etc.). These data are recorded in chronological order to form multiple growth time nodes representing different growth stages. The growth time node refers to a representative or critical time point in the growth process of the melon, such as the 10th day after sowing, the 5th day after flowering, etc. Step S320, the multiple growth time nodes are serialized according to the data collection time sequence to obtain a growth time sequence. Specifically, the multiple growth time nodes are arranged in chronological order to form an ordered growth time sequence. This sequence reflects the changes in environmental data during the entire growth cycle of the melon from sowing to harvesting. Step S330, using a long short-term memory network to train and verify the growth characteristic parameters according to the growth time series, and construct a growth prediction model. Specifically, the growth time series is used as input, and the growth characteristic parameters (such as leaf area, fruit volume, plant height, etc.) are used as target outputs. By training the long short-term memory network, the network learns the laws of the melon growth process, thereby constructing a growth prediction model.

[0037] Step S340, multidimensional prediction is performed through the growth prediction model to generate multidimensional growth prediction data, and the multidimensional growth prediction data includes leaf area prediction data, fruit volume prediction data, and plant height prediction data. Specifically, the trained growth prediction model is used to perform multidimensional prediction on the future growth of the melon, and multidimensional growth prediction data including leaf area prediction data, fruit volume prediction data, plant height prediction data, etc. are generated. Step S350, according to the multiple growth cycles, growth prediction is performed based on the leaf area prediction data to obtain the leaf area prediction trend; step S360, according to the multiple growth cycles, growth prediction is performed based on the fruit volume prediction data to obtain the fruit expansion rate prediction trend; step S370, according to the multiple growth cycles, growth prediction is performed based on the plant height prediction data to obtain the overall growth rate prediction trend. Specifically, according to the leaf area prediction data, a prediction trend curve of the leaf area changing over time is drawn. By calculating the fruit volume change rate at adjacent time nodes, the prediction trend of the fruit expansion rate is obtained. Based on the plant height prediction data, a prediction trend curve of plant height changing over time is drawn to reflect the overall growth rate. Among them, the leaf area prediction trend, the fruit expansion rate prediction trend, and the overall growth rate prediction trend refer to the changing trends of melon leaf area, fruit volume, and plant height in the future period of time. Step S380, according to the multiple growth stages, the leaf area prediction trend, the fruit expansion rate prediction trend, and the overall growth rate prediction trend are associated and integrated to obtain the growth prediction trend curve. Specifically, the leaf area prediction trend, the fruit expansion rate prediction trend, and the overall growth rate prediction trend obtained in steps S350-S370 are comprehensively analyzed, and they are associated and integrated together to form a growth prediction trend curve that can fully reflect the future growth of melons. This implementation method trains and verifies the growth time series through a long short-term memory network, can learn the complex laws in the melon growth process, improve the accuracy of growth prediction, and improve the comprehensiveness of growth prediction through multi-dimensional prediction.

[0038] In a possible implementation, the atomized nutrient solution coefficient is mapped to the growth prediction trend curve based on the growth characteristic parameters for supply analysis, and a nutrient solution supply strategy is formulated. Step S300 further includes step S390, mapping the atomized nutrient solution coefficient to the leaf area prediction trend based on the growth characteristic parameters for supply analysis, and generating first supply and demand data. Specifically, the leaf area prediction trend data is obtained, and based on the knowledge base of melon growth, a mapping rule between leaf area and nitrogen demand is set, wherein a larger leaf area means a higher photosynthesis efficiency and nitrogen nutrient demand. According to the leaf area prediction trend, the amount of nitrogen required for each growth stage is calculated by multiplying the leaf area by the nitrogen demand per unit area. The calculated nitrogen demand is converted into an atomized nutrient solution coefficient, i.e., the first supply and demand data, which is used to guide the preparation and supply of nutrient solution. Step S3100, mapping the atomized nutrient solution coefficient to the fruit expansion rate prediction trend based on the growth characteristic parameters for supply analysis, and generating second supply and demand data. Specifically, the fruit expansion rate prediction trend data is obtained, and based on the knowledge base of melon growth, the mapping rules between the fruit expansion rate and the potassium element demand are set. Potassium is essential for fruit expansion and quality improvement. According to the fruit expansion rate prediction trend, the amount of potassium required for each growth stage is calculated, and the calculated potassium demand is converted into an atomized nutrient solution coefficient, i.e., the second supply and demand data. Step S3110, based on the growth characteristic parameters, the atomized nutrient solution coefficient is mapped to the overall growth rate prediction trend for supply analysis to generate the third supply and demand data. Specifically, the overall growth rate prediction trend data is obtained, which includes the predicted value of the plant height. Based on the knowledge base of melon growth, the mapping rules between the overall growth rate and the nutrient solution supply frequency are set, wherein the faster the growth rate, the higher the nutrient solution supply frequency should be. According to the overall growth rate prediction trend, the nutrient solution supply frequency required for each growth stage is calculated. The calculated supply frequency is converted into a specific nutrient solution supply plan, i.e., the third supply and demand data. Step S3120, based on the first supply and demand data, the second supply and demand data, and the third supply and demand data, synchronously adjust in combination with the multiple growth stages, generate a multi-stage supply result, perform dynamic adjustments based on the multi-stage supply result, and formulate the nutrient solution supply strategy. Specifically, the first, second, and third supply and demand data are integrated to form a complete nutrient solution supply and demand table. Based on the integrated supply and demand table, combined with the growth cycle and growth stage of the melon, a multi-stage supply result is generated, and each stage has a specific nutrient solution formula, supply amount, and supply frequency. In the actual growth process, the multi-stage supply results are dynamically adjusted based on the real-time monitored growth data and growth monitoring logs, and the adjustment content includes the nutrient solution formula, supply amount, and supply frequency.According to the adjusted multi-stage supply results, the final nutrient solution supply strategy is formulated to guide the refined supply control of melons. This implementation method accurately calculates the nutrient requirements of melons at different growth stages by mapping the atomized nutrient solution coefficient to the leaf area prediction trend, fruit expansion rate prediction trend and overall growth rate prediction trend, improves the accuracy of supply, and calculates the nutrient requirements of each growth stage and rationally prepares the nutrient solution, avoiding nutrient waste and oversupply and optimizing resource utilization.

[0039] Step S400, execute the nutrient solution supply strategy to monitor the growth of melons in real time, generate a growth monitoring log, dynamically adjust the atomized nutrient solution coefficient according to the growth monitoring log, and generate a second atomized nutrient solution coefficient. Specifically, supply nutrient solution to melons according to the established nutrient solution supply strategy. Use the sensor network to monitor the growth status of melons and the supply of nutrient solution in real time, record the growth status of melons, the supply of nutrient solution, changes in the growth environment and other information, and generate a growth monitoring log. According to the growth monitoring log, analyze the growth of melons and the supply effect of nutrient solution, dynamically adjust the atomized nutrient solution coefficient, and generate a second atomized nutrient solution coefficient.

[0040] In a possible implementation, the nutrient solution supply strategy is executed to monitor the growth of melons in real time, a growth monitoring log is generated, the atomized nutrient solution coefficient is dynamically adjusted according to the growth monitoring log, and a second atomized nutrient solution coefficient is generated. Step S400 further includes step S410, executing the nutrient solution supply strategy to monitor the growth environment in real time and obtain real-time growth environment change data. Specifically, the environmental parameters such as temperature, humidity, and light intensity of the melon growth area are collected in real time through the sensor network to form real-time growth environment change data. Step S420, executing the nutrient solution supply strategy to monitor the growth characteristics in real time and obtain real-time growth characteristics change data. Specifically, according to the characteristics of melon growth, leaf color, leaf morphology, fruit volume, etc. are selected as indicators for real-time growth characteristics monitoring. Through methods such as image recognition, the growth characteristics of melons are monitored in real time, such as changes in leaf color, growth in fruit volume, etc., to form real-time growth characteristics change data. Step S430, based on the multiple growth cycles, a monitoring time interval is determined, and growth records are recorded according to the monitoring time interval combined with the real-time growth environment change data and the real-time growth characteristics change data to obtain a growth monitoring log. Specifically, according to the growth cycle and growth characteristics of the melon, the monitoring time interval is determined, such as monitoring once a day or several times a week. In each monitoring time interval, the real-time growth environment change data and the real-time growth characteristic change data are integrated to form a growth record. The growth records in each monitoring time interval are summarized and sorted to form a growth monitoring log. Step S440, the growth monitoring log is covered and traversed to generate a coverage traversal result, and anomaly detection is performed according to the coverage traversal result to generate multiple abnormal labels. Specifically, the growth monitoring log is traversed to ensure that each data point is checked. According to the preset anomaly detection rules, the traversed data is detected for anomalies, such as abnormally high or low temperature, abnormal leaf color, etc. The detected abnormal data is marked and an abnormal label is generated. Step S450, the multiple abnormal labels are mapped to the growth monitoring log for matching, multiple abnormal growth data are determined, and the atomized nutrient solution coefficient is dynamically adjusted according to the multiple abnormal growth data to generate the second atomized nutrient solution coefficient. Specifically, the generated abnormal label is matched with the growth monitoring log to determine the specific abnormal growth data. According to the abnormal growth data, the causes of the abnormality are analyzed, such as improper nutrient solution ratio, abnormal environmental parameters, etc. According to the analysis results, the atomized nutrient solution coefficient is dynamically adjusted to generate a second atomized nutrient solution coefficient to optimize the nutrient solution supply strategy. This implementation method can timely detect abnormal conditions in the melon growth process by overwriting and traversing the growth monitoring log and detecting anomalies, providing a basis for timely adjusting the nutrient solution supply strategy.

[0041] In a possible implementation, the multiple abnormal labels are mapped to the growth monitoring log for matching, multiple abnormal growth data are determined, the atomized nutrient solution coefficient is dynamically adjusted according to the multiple abnormal growth data, and the second atomized nutrient solution coefficient is generated. Step S450 further includes step S451, using the multiple abnormal labels as indexes, traversing the growth monitoring log for abnormality determination, and judging whether the growth monitoring log is not in a preset growth interval value. Specifically, the generated multiple abnormal labels are used as indexes, and these labels represent abnormal situations that may occur in the growth monitoring log. Check each data in the growth monitoring log one by one according to the index to ensure that there is no omission. Compare each data with the preset growth interval value, wherein the preset growth interval value is set according to the standard range of normal melon growth, including key indicators such as temperature, humidity, light intensity, leaf area, fruit size, and growth rate. If a certain data exceeds the preset growth interval value, it is recorded as an abnormality. Step S452, if the growth monitoring log contains data that are not in the preset growth interval value, then the growth monitoring data that are not in the preset growth interval value are extracted, and the growth monitoring data that are not in the preset growth interval value are matched with the multiple abnormal labels, and multiple abnormal growth data are determined according to the matching results. Specifically, all data that are not in the preset growth interval value are extracted from the growth monitoring log. These abnormal data are matched with multiple abnormal labels to find out the corresponding abnormal situations. According to the matching results, it is determined which growth data are abnormal, that is, the abnormal growth data is the data in the growth monitoring log that is not in the preset growth interval value and matches the abnormal label.

[0042] Step S453, based on the multiple abnormal growth data combined with the growth prediction trend curve, leaf area analysis is performed to obtain abnormal leaf area data. Specifically, data related to leaf area is extracted from the abnormal growth data, these data are compared with the growth prediction trend curve, and the changing trend of leaf area is analyzed. According to the comparison result, it is determined whether the leaf area is abnormal, and the abnormal data (the leaf area does not match the growth prediction trend curve and exceeds the normal range) is recorded. Step S454, based on the multiple abnormal growth data combined with the growth prediction trend curve, fruit enlargement analysis is performed to obtain abnormal fruit enlargement data. Specifically, data related to fruit enlargement is extracted from the abnormal growth data, these data are compared with the growth prediction trend curve, and the changing trend of fruit enlargement is analyzed. According to the comparison result, it is determined whether the fruit enlargement is abnormal, and the abnormal data (the fruit enlargement does not match the growth prediction trend curve and exceeds the normal range) is recorded. Step S455, based on the multiple abnormal growth data combined with the growth prediction trend curve, overall growth analysis is performed to obtain abnormal growth rate data. Specifically, data related to overall growth are extracted from the abnormal growth data, and these data are compared with the growth prediction trend curve to analyze the changing trend of overall growth. According to the comparison result, it is determined whether the growth rate is abnormal, and the abnormal data (the overall growth does not conform to the growth prediction trend curve and exceeds the normal range) is recorded. Step S456, the leaf area abnormal data, the fruit enlargement abnormal data, and the growth rate abnormal data are integrated, and the atomized nutrient solution coefficient is adjusted and updated according to the integration result to obtain the second atomized nutrient solution coefficient. Specifically, the leaf area abnormal data, the fruit enlargement abnormal data, and the growth rate abnormal data are integrated to form a complete abnormal growth data set. According to the abnormal growth data set, the causes of the abnormality are analyzed, such as improper nutrient solution ratio, abnormal environmental parameters, etc. According to the analysis results, the atomized nutrient solution coefficient is adjusted and updated to optimize the supply strategy of the nutrient solution. This implementation method deeply analyzes the causes of the abnormality by integrating the leaf area abnormal data, the fruit enlargement abnormal data, and the growth rate abnormal data, provides a basis for adjusting the nutrient solution supply strategy, and optimizes the supply strategy of the nutrient solution.

[0043] Step S500, based on the second atomized nutrient solution coefficient, the nutrient solution supply strategy is updated and optimized, and a nutrient solution supply optimization strategy is generated to perform adaptive and refined supply control of the nutrient solution for the mist culture melon. Specifically, based on the second atomized nutrient solution coefficient, the original nutrient solution supply strategy is updated and optimized, and a nutrient solution supply optimization strategy is generated. According to the nutrient solution supply optimization strategy, the melon is adaptively and refinedly supplied, that is, the supply method and supply amount of the nutrient solution are automatically adjusted according to the growth status and nutritional needs of the melon, to ensure that the melon can obtain just the right nutritional support at different growth stages. The embodiment of the present application collects environmental data of the melon growing area in real time through a sensor network, performs growth analysis on the melon, calculates nutritional requirements based on the analysis results, determines the atomized nutrient solution coefficient, and at the same time, constructs a growth prediction model, synchronizes growth characteristic parameters to the model, obtains a growth prediction trend curve, and performs supply analysis based on the curve to formulate a nutrient solution supply strategy. During the execution process, the growth of the melon is monitored in real time, a growth monitoring log is generated, and the atomized nutrient solution coefficient is dynamically adjusted according to the log, and other technical means are used to achieve the technical effect of realizing refined control through real-time, accurate and dynamic adjustment of the nutrient solution.

[0044] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for finely controlling the supply of nutrient solution based on the prediction of the growth of aeroponic melons, characterized in that: The method comprises: The sensor network is used to collect data on the melon growing area to obtain the growing environment data; Performing growth analysis on the melon according to the growth environment data to obtain growth characteristic parameters, calculating the nutrient demand according to the growth characteristic parameters, and determining the atomized nutrient solution coefficient; Constructing a growth prediction model, synchronizing the growth characteristic parameters to the growth prediction model according to the growth environment data, obtaining a growth prediction trend curve, mapping the atomized nutrient solution coefficient to the growth prediction trend curve based on the growth characteristic parameters to perform supply analysis, and formulating a nutrient solution supply strategy; Executing the nutrient solution supply strategy to monitor the growth of the melon in real time, generating a growth monitoring log, dynamically adjusting the atomized nutrient solution coefficient according to the growth monitoring log, and generating a second atomized nutrient solution coefficient; The nutrient solution supply strategy is updated and optimized based on the second atomized nutrient solution coefficient, and a nutrient solution supply optimization strategy is generated to perform adaptive and refined supply control of the nutrient solution for the atomized melon.

2. The method for fine supply control of nutrient solution based on growth prediction of aeroponic melon according to claim 1, characterized in that: Performing growth analysis on the melon according to the growth environment data to obtain growth characteristic parameters comprises: Retrieving a historical growth data record log of melons in the melon growth area, and matching the historical growth data record log based on the growth environment data to obtain a full growth fluctuation data set of the melons, wherein the full growth fluctuation data set includes multiple growth stages and multiple growth cycles; Perform growth detection by traversing the multiple growth stages according to the multiple growth cycles, generate detection results, and determine the real-time growth stage of the melon according to the detection results; Perform growth impact analysis on the melon based on the real-time growth stage to obtain multiple impact factors; The melon growth is evaluated according to the multiple influencing factors to obtain a growth impact evaluation result, and the growth characteristic parameters are obtained by performing a growth characteristic analysis on the growth impact evaluation result based on the multiple growth cycles.

3. The method for controlling the refined supply of nutrient solution based on the prediction of the growth of aeroponic melon according to claim 2, characterized in that: The nutrient requirement is calculated according to the growth characteristic parameters to determine the atomized nutrient solution coefficient, and the method includes: Performing demand analysis according to the real-time growth stage combined with the growth characteristic parameters to determine a plurality of nutrient demand elements, wherein the plurality of nutrient demand elements include atomization frequency of nutrient solution; Based on the multiple nutritional requirement elements, weight allocation is performed according to the growth impact assessment result to obtain multiple weight coefficients; Calculate the nutrient requirement according to the nutrient solution atomization frequency in combination with the multiple weight coefficients to obtain a concentration coefficient of the atomized nutrient solution; The nutrient requirement of the melon is determined based on the concentration coefficient of the atomized nutrient solution in combination with the real-time growth stage to determine the atomized nutrient solution coefficient.

4. The method for controlling the refined supply of nutrient solution based on the prediction of the growth of aeroponic melon according to claim 3, characterized in that: The method comprises: determining the nutrient requirement of the melon based on the concentration coefficient of the atomized nutrient solution and combining the real-time growth stage to determine the coefficient of the atomized nutrient solution. Performing simulated nutrient supply based on the atomized nutrient solution concentration coefficient to generate a simulated nutrient supply result; According to the simulated nutrient supply result, judging whether the real-time growth state of the melon in the real-time growth stage meets the expected growth trend value, the real-time growth stage including the real-time growth state; If the real-time growth state of the melon in the real-time growth stage does not meet the expected growth trend value, an adjustment instruction is generated, and the nutrient solution atomization frequency is dynamically adjusted through the adjustment instruction to obtain the nutrient solution atomization adjustment frequency; The concentration coefficient of the atomized nutrient solution is updated and calculated according to the atomization adjustment frequency of the nutrient solution to obtain the atomized nutrient solution coefficient; If the real-time growth state of the melon in the real-time growth stage meets the expected growth trend value, a control instruction is generated, and the atomization of the nutrient solution is controlled according to the atomized nutrient solution concentration coefficient through the control instruction to obtain the atomized nutrient solution coefficient.

5. The method for controlling the refined supply of nutrient solution based on the growth prediction of aeroponic melon according to claim 2, characterized in that: Constructing a growth prediction model, synchronizing the growth characteristic parameters to the growth prediction model according to the growth environment data, and obtaining a growth prediction trend curve, the method comprising: Based on the growth environment data, the melon is collected in a time series to obtain a plurality of growth time nodes; Performing serial processing on the multiple growth time nodes according to the data collection time sequence to obtain a growth time sequence; Using a long short-term memory network to train and verify the growth characteristic parameters according to the growth time series, and construct a growth prediction model; Performing multidimensional prediction through the growth prediction model to generate multidimensional growth prediction data, wherein the multidimensional growth prediction data includes leaf area prediction data, fruit volume prediction data, and plant height prediction data; According to the plurality of growth cycles, growth prediction is performed based on the leaf area prediction data to obtain a leaf area prediction trend; According to the multiple growth cycles, growth prediction is performed based on the fruit volume prediction data to obtain a fruit expansion rate prediction trend; According to the multiple growth cycles, growth prediction is performed based on the plant height prediction data to obtain an overall growth rate prediction trend; According to the multiple growth stages, the leaf area prediction trend, the fruit expansion rate prediction trend, and the overall growth rate prediction trend are correlated and integrated to obtain the growth prediction trend curve.

6. The method for controlling the refined supply of nutrient solution based on the growth prediction of aeroponic melon according to claim 5, characterized in that: Based on the growth characteristic parameters, the atomized nutrient solution coefficient is mapped to the growth prediction trend curve to perform supply analysis and formulate a nutrient solution supply strategy, the method comprising: Based on the growth characteristic parameters, the atomized nutrient solution coefficient is mapped to the leaf area prediction trend to perform supply analysis to generate first supply demand data; Based on the growth characteristic parameters, the atomized nutrient solution coefficient is mapped to the fruit expansion rate prediction trend to perform supply analysis and generate second supply and demand data; Based on the growth characteristic parameters, the atomized nutrient solution coefficient is mapped to the overall growth rate prediction trend to perform supply analysis to generate third supply demand data; Based on the first supply and demand data, the second supply and demand data, and the third supply and demand data, synchronous adjustments are made in combination with the multiple growth stages to generate multi-stage supply results, and dynamic adjustments are made according to the multi-stage supply results to formulate the nutrient solution supply strategy.

7. The method for controlling the refined supply of nutrient solution based on the prediction of the growth of aeroponic melon according to claim 2, characterized in that: The nutrient solution supply strategy is executed to monitor the growth of the melon in real time, a growth monitoring log is generated, and the atomized nutrient solution coefficient is dynamically adjusted according to the growth monitoring log to generate a second atomized nutrient solution coefficient. The method includes: Execute the nutrient solution supply strategy to perform real-time growth environment monitoring and obtain real-time growth environment change data; Executing the nutrient solution supply strategy to monitor real-time growth characteristics and obtain real-time growth characteristic change data; Determine a monitoring time interval based on the multiple growth cycles, and record the growth according to the monitoring time interval in combination with the real-time growth environment change data and the real-time growth characteristic change data to obtain a growth monitoring log; Performing a coverage traversal on the growth monitoring log to generate a coverage traversal result, performing anomaly detection according to the coverage traversal result, and generating multiple anomaly labels; The multiple abnormal labels are mapped to the growth monitoring log for matching, a plurality of abnormal growth data are determined, the atomized nutrient solution coefficient is dynamically adjusted according to the multiple abnormal growth data, and the second atomized nutrient solution coefficient is generated.

8. The method for controlling the refined supply of nutrient solution based on the growth prediction of aeroponic melon according to claim 7, characterized in that: The method includes mapping the multiple abnormal labels to the growth monitoring log for matching, determining multiple abnormal growth data, dynamically adjusting the atomized nutrient solution coefficient according to the multiple abnormal growth data, and generating the second atomized nutrient solution coefficient. Using the multiple abnormal tags as indexes, traversing the growth monitoring log to perform abnormality determination, and determining whether there is a value in the growth monitoring log that is not within a preset growth interval; If the growth monitoring log contains data that are not in the preset growth interval, extracting the growth monitoring data that are not in the preset growth interval, matching the growth monitoring data that are not in the preset growth interval with the multiple abnormal labels, and determining multiple abnormal growth data according to the matching results; Perform leaf area analysis based on the plurality of abnormal growth data in combination with the growth prediction trend curve to obtain leaf area abnormality data; Performing fruit expansion analysis based on the plurality of abnormal growth data in combination with the growth prediction trend curve to obtain abnormal fruit expansion data; Performing overall growth analysis based on the plurality of abnormal growth data in combination with the growth prediction trend curve to obtain abnormal growth rate data; The leaf area abnormality data, the fruit enlargement abnormality data, and the growth rate abnormality data are integrated, and the atomized nutrient solution coefficient is adjusted and updated according to the integration result to obtain the second atomized nutrient solution coefficient.

Images