Fertilization recommendation method, system and electronic equipment

Abstract

The present invention provides a method, system and electronic device for recommending fertilizers, in the technical field of fertilizer recommendation, the method is applied to a fertilizer recommendation platform based on a database, the database is used to store the standard leaf contents of multiple nutrient elements required by fruit trees in multiple orchards; the method comprises: obtaining the leaf contents of multiple nutrient elements in the target orchard of the fruit growth period; retrieving the standard leaf contents of multiple nutrient elements in the target orchard from the database; determining the recommended fertilizer plan for the target orchard according to the multiple standard leaf contents and the multiple leaf contents. The present invention determines the recommended fertilizer plan for the target orchard based on the leaf contents of nutrient elements in the target orchard of the fruit growth period, thereby improving the rationality of the recommended fertilizer plan.

Description

Technical Field

[0001] The present invention relates to the technical field of fertilization recommendation, and in particular to a fertilization recommendation method, system and electronic equipment. Background Art

[0002] China ranks first in the world in terms of citrus planting area and output. Market demand has gradually shifted from quantity to quality, and the quality of citrus fruit has become the key to determining market competitiveness. Studies have shown that appropriate fertilization is an important factor affecting citrus quality, but due to the diversity of citrus varieties, ecological climate, soil fertility and soil types in citrus production areas, it is difficult for citrus farmers to accurately fertilize according to the nutritional status of the trees. Liang Shanshan's survey results in 2017 showed that citrus orchards that applied excessive nitrogen (N) fertilizers, phosphorus (P) fertilizers and potassium (K) fertilizers accounted for 57.5%, 77.0% and 69.4% of China's total citrus planting area, respectively, and citrus orchards that applied organic fertilizers accounted for 47.8%. Excessive application of chemical fertilizers and excessive replacement of chemical fertilizers with organic fertilizers in the same production area will lead to deterioration of fruit quality. Therefore, it is urgent to promote the technology of formula fertilization in citrus orchards to improve fruit quality and maintain the sustainable development of the citrus industry.

[0003] In the past 20 years, crop fertilization formula methods have mostly been supported by soil testing and formula technology, but the soil sampling and testing methods involved in soil testing and formula technology are uncertain, and it is difficult to use to establish nutritional standards, which will consume a lot of manpower and time costs. At the same time, the roots of fruit crops such as citrus are widely distributed in the soil, and it is more difficult to obtain representative soil samples than field crops, resulting in unreasonable recommendation results. Summary of the invention

[0004] The purpose of the present invention is to provide a fertilization recommendation method, system and electronic equipment, which determine the fertilization recommendation plan of the target orchard based on the leaf content of nutrient elements in the target orchard during the fruit growth period, thereby improving the rationality of the fertilization recommendation plan.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A fertilization recommendation method, the fertilization recommendation method is applied to a database-based fertilization recommendation platform, the fertilization recommendation platform includes a system server and multiple user terminals;

[0007] The system server is connected to the plurality of user terminals respectively;

[0008] The user terminal is used to obtain leaf contents of multiple nutrient elements required by fruit trees in multiple orchards, and fruit quality in multiple orchards;

[0009] The system server is used to construct a database based on the leaf contents of multiple nutrient elements required by fruit trees in multiple orchards and the fruit quality of multiple orchards; the database is used to store the standard leaf contents of multiple nutrient elements required by fruit trees in multiple orchards;

[0010] The user terminal is used to obtain a target orchard fertilization recommendation plan request;

[0011] The system server is used to generate a fertilization recommendation plan for the target orchard according to the fertilization recommendation plan request of the target orchard;

[0012] The user terminal is also used to display the recommended fertilization plan;

[0013] The method comprises:

[0014] Obtain leaf contents of multiple nutrients in the fruit Pengda period target orchard;

[0015] The standard leaf contents of multiple nutrient elements in the target orchard are retrieved from the database;

[0016] A recommended fertilization plan for a target orchard is determined based on the plurality of standard leaf contents and the plurality of leaf contents.

[0017] Optionally, the step of retrieving the standard leaf contents of multiple nutrient elements of the target orchard from the database includes:

[0018] Determine whether the database stores standard leaf contents of multiple nutrient elements required by fruit trees in the target orchard, and obtain a first determination result;

[0019] If the first judgment result is yes, then the standard leaf contents of multiple nutrient elements of the target orchard are retrieved from the database;

[0020] If the first judgment result is no, determining the orchard closest to the target orchard as the target replacement orchard;

[0021] The standard leaf contents of various nutrients in the target replacement orchard are retrieved as the standard leaf contents of various nutrients in the target orchard.

[0022] Optionally, before obtaining the leaf contents of multiple nutrient elements in the fruit growth period target orchard, it also includes:

[0023] The database is constructed.

[0024] Optionally, the constructing the database includes:

[0025] Determine any orchard as the current orchard;

[0026] Get the current orchard fruit quality;

[0027] Obtain the leaf content of multiple nutrients in the current orchard of Fruit Peng Daqi;

[0028] According to the current orchard fruit quality and leaf contents of multiple nutrients, the nutrient component analysis method was used to determine the proposed standard leaf content of each nutrient element.

[0029] Optionally, obtaining the current orchard fruit quality includes:

[0030] Obtain multiple fruit quality indicators of multiple sample fruits in the current orchard during the fruit maturity period;

[0031] Determine the sum of the products of each fruit quality index and the corresponding weight as the sample fruit quality;

[0032] The average value of the sample fruit quality was determined as the current orchard fruit quality.

[0033] Optionally, the obtaining of the current orchard fruit quality and determining a recommended fertilization scheme for the target orchard according to the plurality of standard leaf contents and the plurality of leaf contents include:

[0034] Determine any nutrient element as the current nutrient element;

[0035] According to the standard leaf content and leaf content of the current nutrient elements, determine whether the current nutrient elements are the nutrient elements that need to be applied to the target orchard;

[0036] When the current nutrient element is the target orchard that needs to apply the nutrient element, the current correction coefficient is determined using the formula ΔC=1-(C-CCV) / CCV according to the standard leaf content and leaf content of the current nutrient element; wherein ΔC is the current correction coefficient; C is the leaf content of the current nutrient element; and CCV is the standard leaf content of the current nutrient element.

[0037] According to the current correction coefficient, the formula R = U × Y × ΔC is used to determine the fertilizer requirement of the current nutrient element; U is the nutrient absorption of the fruit tree for the current nutrient element during the fruit growth period; Y is the target fruit yield;

[0038] Update the current nutrient element, and return to the step of "determining whether the current nutrient element is the nutrient element that needs to be applied to the target orchard according to the standard leaf content and leaf content of the current nutrient element";

[0039] Determine the recommended fertilization plan for the target orchard based on the nutrients to be applied and the required amount of nutrients to be applied.

[0040] Optionally, determining whether the current nutrient element is a nutrient element that needs to be applied to the target orchard based on the standard leaf content and leaf content of the current nutrient element includes:

[0041] Determine whether the absolute value of the difference between the standard leaf content and the leaf content of the current nutrient element is less than a preset difference threshold, and obtain a second determination result;

[0042] If the second judgment result is yes, then determine that the current nutrient element does not need to be added; update the current nutrient element,

[0043] If the second judgment result is no, it is determined that the current nutrient element is a nutrient element that needs to be added.

[0044] A fertilization recommendation system, comprising:

[0045] Leaf content acquisition module, used to obtain the leaf content of multiple nutrient elements in the target orchard during the fruit growth period;

[0046] A standard leaf content retrieval module is used to retrieve the standard leaf content of multiple nutrient elements in the target orchard from the database;

[0047] The fertilization recommendation scheme determination module is used to determine the fertilization recommendation scheme of the target orchard based on the multiple standard leaf contents and the multiple leaf contents.

[0048] An electronic device comprises a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the fertilization recommendation method.

[0049] Optionally, the memory is a readable storage medium.

[0050] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0051] The present invention provides a fertilization recommendation method, system and electronic device. The method is applied to a fertilization recommendation platform based on a database, wherein the database is used to store the standard leaf contents of multiple nutrient elements required by fruit trees in multiple orchards; the method comprises: obtaining the leaf contents of multiple nutrient elements in a target orchard of fruit growth; retrieving the standard leaf contents of multiple nutrient elements in the target orchard from the database; and determining the fertilization recommendation scheme of the target orchard according to multiple standard leaf contents and multiple leaf contents. The present invention determines the fertilization recommendation scheme of the target orchard based on the leaf contents of nutrient elements in the target orchard of fruit growth, thereby improving the rationality of the fertilization recommendation scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0053] Figure 1 This is a flow chart of a recommended fertilization method in Example 1 of the present invention;

[0054] Figure 2 This is a schematic diagram of the agricultural industrial park in the test site in Example 1 of the present invention;

[0055] Figure 3 This is a decision flow chart of a fertilization recommendation platform based on a database in Example 1 of the present invention;

[0056] Figure 4 The overall structure diagram of the database-based fertilization recommendation platform in Example 1 of the present invention;

[0057] Figure 5 This is a schematic diagram of a fertilization recommendation page of a fertilization recommendation platform based on a database in Example 1 of the present invention;

[0058] Figure 6 This is a diagram of the upload interface for the original data of fruit quality in Example 1 of the present invention;

[0059] Figure 7 This is a diagram of the upload interface for the original leaf mineral nutrient data in Example 1 of the present invention;

[0060] Figure 8 This is a diagram showing the interface for viewing the original data of the production area and the suitable standard of leaf nutrients in Example 1 of the present invention;

[0061] Fig. 9 This is a diagram of the nutritional balance status evaluation interface of an orange orchard in a production area in Example 1 of the present invention;

[0062] Fig.10 This is a page diagram for recommending compound fertilizer solutions in Example 1 of the present invention;

[0063] Figure 11 is a comparison diagram of the intrinsic quality of Longan pomelo fruit under different treatments in Example 1 of the present invention; Figure 11(A) is a comparison diagram of the soluble solid content of Longan pomelo fruit under different treatments in Example 1 of the present invention; Figure 11(B) is a comparison diagram of the titratable acid content of Longan pomelo fruit under different treatments in Example 1 of the present invention; Figure 11(C) is a comparison diagram of the maturity index of Longan pomelo fruit under different treatments in Example 1 of the present invention; Figure 11(D) is a comparison diagram of the vitamin C content of Longan pomelo fruit under different treatments in Example 1 of the present invention; Figure 11(E) is a comparison diagram of the juice yield of Longan pomelo fruit under different treatments in Example 1 of the present invention; Figure 11(F) is a comparison diagram of the edible rate of Longan pomelo fruit under different treatments in Example 1 of the present invention;

[0064] Fig.12 This is a comparison diagram of the appearance of the fruit and the color of the flesh under different treatments in Example 1 of the present invention;

[0065] Fig.13 This is a comparison chart of the comprehensive fruit quality index of Longan pomelo under different treatments in Example 1 of the present invention;

[0066] Fig.14 This is a diagram of the execution result interface of the compound fertilizer formula recommendation algorithm in Example 1 of the present invention. DETAILED DESCRIPTION

[0067] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0068] The purpose of the present invention is to provide a fertilization recommendation method, system and electronic equipment, which determine the fertilization recommendation plan of the target orchard based on the leaf content of nutrient elements in the target orchard during the fruit growth period, thereby improving the rationality of the fertilization recommendation plan.

[0069] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0070] The tree nutrient status of adult fruit crops is stable, and the leaf mineral nutrition diagnosis can reflect the fertility status of the orchard. Technicians have conducted a large number of fruit tree leaf mineral nutrient diagnosis studies for the fruit expansion period, and established citrus formula recommended fertilization technical regulations in various places. Among them, the establishment of suitable standards for leaf mineral nutrients is a prerequisite for realizing the diagnosis of the nutritional status of orange orchards and recommending fertilization plans. In 2001, Khiari et al. established the relationship between the cumulative variance function of leaf nutrient elements and yield or fruit quality through the nutrient component analysis method (CND, Component Nutrition Diagnosis Method), obtained the inflection point value through two derivations, realized the establishment of suitable standards for mineral nutrition of crop leaves and the evaluation of nutritional balance status, and paid more attention to the representativeness of yield in the data set; Martin et al. used the Deviation from Optimum Percentage (DOP) method to realize the direct measurement of the limiting order of nutrient elements and the deviation of each element content in low-level orange orchards. The above two types of nutrition diagnosis algorithms have been applied to fruit crops such as apples, lychees, and navel oranges. Most of the above diagnostic methods use yield as the classification standard, but the fruit expansion period is a critical period for balancing nutrient input and improving fruit quality. The nutritional status of leaves during this period is more related to the comprehensive quality of the fruit. The formula fertilization recommendation technology established with yield as the goal has certain limitations in guiding fertilization during the citrus fruit expansion period. At the same time, compared with the current mature crop nutrition fertilization recommendation expert system, due to the confidentiality of the data set and algorithm library, there is no citrus formula fertilization recommendation system for growers that can be promoted and used. To this end, the present invention provides the following fertilization recommendation method, system and electronic device.

[0071] Example 1

[0072] like Figure 1 This embodiment provides a fertilization recommendation method, which is applied to a database-based fertilization recommendation platform, and the fertilization recommendation platform includes a system server and multiple user terminals; the system server is connected to the multiple user terminals respectively; the user terminals are used to obtain the leaf content of multiple nutrients required by fruit trees in multiple orchards, and the fruit quality of multiple orchards; the system server is used to build a database according to the leaf content of multiple nutrients required by fruit trees in multiple orchards, and the fruit quality of multiple orchards; the database is used to store the standard leaf content of multiple nutrients required by fruit trees in multiple orchards; the user terminal is used to obtain a target orchard fertilization recommendation plan request; the system server is used to generate a target orchard fertilization recommendation plan according to the target orchard fertilization recommendation plan request; the user terminal is also used to display the fertilization recommendation plan.

[0073] Recommended fertilization methods include:

[0074] Step 101: Obtain leaf contents of multiple nutrient elements in the fruit Peng Daqi target orchard.

[0075] Step 102: retrieve the standard leaf contents of multiple nutrient elements of the target orchard from the database.

[0076] Step 102 includes:

[0077] Step 1021: Determine whether the database stores standard leaf contents of multiple nutrient elements required by fruit trees in the target orchard, and obtain a first determination result; if the first determination result is yes, execute step 1022; if the first determination result is no, execute step 1023.

[0078] Step 1022: retrieve the standard leaf contents of multiple nutrient elements of the target orchard from the database.

[0079] Step 1023: Determine the orchard closest to the target orchard as the target replacement orchard.

[0080] Step 1024: retrieve the standard leaf contents of multiple nutrient elements of the target replacement orchard as the standard leaf contents of multiple nutrient elements of the target orchard.

[0081] Step 103: Determine a recommended fertilization plan for the target orchard based on the multiple standard leaf contents and the multiple leaf contents.

[0082] Step 103 includes:

[0083] Step 1031: Determine any nutrient element as the current nutrient element;

[0084] Step 1032: determining whether the current nutrient element is a nutrient element that needs to be applied to the target orchard based on the standard leaf content and leaf content of the current nutrient element;

[0085] Step 1032 includes:

[0086] Step 10321: Determine whether the absolute value of the difference between the standard leaf content and the leaf content of the current nutrient element is less than a preset difference threshold, and obtain a second judgment result; if the second judgment result is yes, execute step 10322; if the second judgment result is no, execute step 10323.

[0087] Step 10322: Determine that the current nutrient element is a nutrient element that does not need to be applied; update the current nutrient element.

[0088] Step 10323: Determine that the current nutrient element is the nutrient element that needs to be applied.

[0089] Step 1033: when the current nutrient element is the target orchard to be applied nutrient element, the current correction coefficient is determined according to the standard leaf content and leaf content of the current nutrient element using the formula ΔC=1-(C-CCV) / CCV; wherein ΔC is the current correction coefficient; C is the leaf content of the current nutrient element; and CCV is the standard leaf content of the current nutrient element.

[0090] Step 1034: according to the current correction coefficient, the fertilizer requirement of the current nutrient element is determined using the formula R=U×Y×ΔC; wherein R is the fertilizer requirement of the current nutrient element; U is the nutrient absorption of the fruit tree for the current nutrient element during the fruit growth period; and Y is the target fruit yield;

[0091] Step 1035: update the current nutrient element, and return to the step of "determining whether the current nutrient element is a nutrient element that needs to be applied to the target orchard according to the standard leaf content and leaf content of the current nutrient element";

[0092] Step 1036: Determine a recommended fertilization plan for the target orchard based on the nutrients to be applied and the required amount of the nutrients to be applied.

[0093] Before step 101, the method further includes:

[0094] Step 104: Build a database.

[0095] Step 104 includes:

[0096] Step 1041: Determine any orchard as the current orchard.

[0097] Step 1042: Obtain the current orchard fruit quality.

[0098] Step 1042 includes:

[0099] Step 10421: Obtain multiple fruit quality indicators of multiple sample fruits in the current orchard during the fruit maturity period.

[0100] Step 10422: Determine the sum of the products of each fruit quality index and the corresponding weight as the sample fruit quality.

[0101] Step 10423: Determine the average value of the sample fruit quality as the current orchard fruit quality.

[0102] Step 1043: Obtain the leaf contents of various nutrient elements in the fruit garden at the current time.

[0103] Step 1044: Based on the current orchard fruit quality and the leaf contents of multiple nutrient elements, the to-be-determined standard leaf content of each nutrient element is determined using a nutrient component analysis method.

[0104] Below, taking Guang'an Longan pomelo as an example, the above technical solution provided by the present invention is specifically described.

[0105] 1. Materials and Methods

[0106] 1.1 Data Source

[0107] like Figure 2 The data in this example are from the Guang'an Long'an Pomelo Modern Agricultural Industrial Park, which is distributed in the Long'an Township Orange Garden (CF) where chemical fertilizers are mainly used and the Chongwang Township Orange Garden (OF) where organic fertilizers are mainly used. In winter, 5kg / plant (kilogram per plant) of commercial organic fertilizer and 3kg of compound fertilizer (26:10:16) were applied as basal fertilizer in the Long'an production area; 20kg / plant of decomposed organic fertilizer (3.63% total nitrogen, 3.01% total potassium, 0.59% total phosphorus) was applied as basal fertilizer in the Chongwang production area, and no chemical fertilizer was applied.

[0108] 1.1.1 Algorithm construction data

[0109] Each township selected 20 plots with basically consistent growth and strong trees, representing the local management level, for a total of 40 plots. In order to avoid the impact of alternate years, the mineral nutrition of spring shoot leaves during the fruit expansion period (early June) and the fruit quality indicators during the maturity period (early November) were tracked at fixed points for two consecutive years in 2019-2020, which were used to construct algorithm libraries for the generation of suitable standards for high-quality leaf nutrients, the evaluation of nutrient status in orange orchards, and the correction of effective nutrient prescriptions.

[0110] 1.1.2 Effect Verification Data

[0111] In 2021, according to the 2021 data sets of Long'an Township Orange Orchard (CF) and Chongwang Township Orange Orchard (OF), the citrus nutrient fertilization recommendation system generated 6 treatments marked as CRF1, CRF2, CRF3 and ORF1, ORF2, ORF3, respectively, which were applied during the fruit expansion period (mid-June), and the empirical fertilization treatment (CFP) of Long'an Township Orange Orchard and the empirical fertilization treatment (OFP) of Chongwang Township Orange Orchard were used as controls, totaling 8 treatments. Five plots were randomly selected for each treatment, for a total of 40 plots. Fixed-point tracking of mineral nutrition and fruit quality indicators of spring shoot leaves at maturity was used to verify the effectiveness of the citrus nutrient fertilization recommendation system.

[0112] 1.2 Index determination

[0113] Five plants were randomly selected from each plot using the "Z" shape method. For leaf samples, 8 healthy and mature spring shoot leaves were collected from each plant around the middle of the crown, totaling 40 leaf mixed samples / plot. For fruit samples, 4 fruits were picked from each plant around the middle of the crown during the fruit ripening period, totaling 20 fruit mixed samples / plot.

[0114] 1.2.1 Leaf mineral nutrition

[0115] The leaf nutrient index used the plant nutrient determination method proposed by Bao Shidan, the Kjeldahl method was used to determine the total N content in the sample, the molybdenum antimony colorimetric method was used to determine the total P content in the sample, the flame photometry was used to determine the total K content in the sample, and the plasma spectrometer (ICP) was used to determine calcium (Ca), magnesium (Mg), iron (Fe), manganese (Mn), copper (Cu) and zinc (Zn).

[0116] 1.2.2 Fruit quality

[0117] Fruit indexes: single fruit weight (PFW) was measured by electronic analytical balance; vertical diameter (HD) and horizontal diameter (VD) of the fruit were measured by vernier caliper, and the fruit shape index was calculated by vertical diameter / horizontal diameter; the color parameters of fruit appearance L* (fruit brightness value), a* (red-green difference) and b* (yellow-blue difference) were measured by Minolta colorimeter (model CR-10), and the fruit comprehensive color index CCI was (1000×a*) / (L*×b*). The internal quality indexes such as fruit soluble solid content (TSS), titratable acid (TA), vitamin C (VC), juice yield (ER) and edible rate (JR) were tested and analyzed with reference to the fresh citrus fruit inspection method (GB / T8210-2011), and the fruit maturity index FMI was calculated as TSS / TA.

[0118] 1.3 Data processing

[0119] Microsoft Excel 2019 was used for processing and the data were expressed as mean ± standard deviation. SPSS 25.0 was used for Duncan (English name, transliterated as Duncan) significance test and principal component analysis, and GraphPad Prism 8.0 was used for drawing.

[0120] 2. Nutritional diagnosis model and fertilization recommendation process

[0121] The recommended process for citrus nutrition diagnosis and fertilization is as follows: Figure 3 As shown in the figure, the system background uploads and updates the basic information and raw data of the orange orchard, and the system server calls and executes the fruit quality comprehensive index algorithm, leaf nutrient suitability standard generation algorithm, orange orchard nutrition status evaluation algorithm, effective nutrient formula correction algorithm and compound fertilizer formula recommendation algorithm (the compound fertilizer formula recommendation algorithm interface is shown in the figure). Fig.14As shown in the figure, it can communicate with the local agricultural resources server in two directions. Users can use the web browser (system front desk) to query their own orange orchard basic information, suitable standards for high-quality mineral nutrition, nutritional status evaluation, effective nutrient formulas and recommended fertilization plans based on compound fertilizer formulas in the local agricultural resources library. Due to the limitations of manpower and material costs for sample collection and determination, the determination cycle for macroelements is usually 2-3 years, and for trace elements is 4-5 years. If there is no mineral nutrition data for the orange orchard fruit expansion period in the current year in the database, the system operation results will be displayed according to the most recent data set.

[0122] 2.1 Algorithm library construction

[0123] 2.1.1 Comprehensive evaluation algorithm for fruit quality

[0124] The comprehensive evaluation algorithm system of fruit quality realizes the standardization of fruit quality data, the principal component analysis and screening of the core indicators of fruit quality, and the automatic assignment of the hierarchical analysis method based on the contribution ranking of each principal component of the core indicators of fruit quality. Finally, the product of the weight coefficient of each indicator and the standardized value is the comprehensive fruit quality score Fq. The higher the score, the better the fruit quality. The classification threshold is: ≧0.60 for first-class fruit, 0.45-0.60 for second-class fruit, 0.30-0.45 for third-class fruit, and less than 0.30 for fourth-class fruit.

[0125] 2.1.2 Algorithm for generating leaf mineral nutrition suitability standards

[0126] According to the CND inflection point value algorithm for high and low yield plots, the yield index was replaced by the comprehensive fruit quality score Fq. The comprehensive fruit quality and leaf mineral nutrition data of the fruit expansion period in the data set were used to fit the three-variable linear equation of the analysis parameters FCi(Vx) and Fq of each mineral nutrient element:

[0127] FCi(Vx)=AFq 3 +BF 2 +CFq+D (1)

[0128] Where: X represents different mineral nutrients, A, B, C and D are all constants. According to formula (3), taking two derivatives, we can get When the comprehensive fruit quality score Fq = -B / 3A, Fq at this time is the inflection point value determined by the algorithm for dividing the high and low quality groups with the X element. The maximum inflection point value with the inflection point value of each mineral element within the data set range and the proportion of high-quality plots in the total data set greater than 12% is selected as the critical value for selecting the high and low quality groups. The leaf mineral nutrition range of the high-quality group is the suitable standard for high-quality leaf mineral nutrition during the fruit expansion period under local conditions.

[0129] 2.1.3 Orange orchard nutritional status diagnosis algorithm

[0130] According to the DOP diagnostic index algorithm

[24] , taking the mean values ​​of each nutrient element in a single orange orchard as the data set, the optimum value percentage deviation Ix and the DOP nutrient imbalance index NBIm were calculated according to formulas (2) and (3):

[0131] Ix=[(C×100) / CCV]-100 (2)

[0132] NBIm=Σ|I X | / n (3)

[0133] Among them: Ix represents the DOP index of N, P, K..., C is the average content of a certain element in the diagnosed orange orchard, CCV is the average value of the suitable standard of high-quality nutrients in the fruit expansion period generated by 1.2.2, NBIm reflects the average value of the sum of the diagnostic indexes of each element. The lower the value, the more balanced the tree nutrients are, and n is the number of diagnosed elements. X |≤NBIm, C within the appropriate standard established by the system indicates that the element content is normal, C below the minimum value of the appropriate standard indicates that it is too low, and C above the maximum value of the appropriate standard indicates that it is too high; |I X |>NBIm, I X When <0, it means the element is lacking, I X >0 indicates an excess of the element.

[0134] 2.1.4 Effective nutrient formula correction algorithm

[0135] Taking the average value of each nutrient in a single orange orchard as the nutrient calculation unit, the effective nutrient demand of nitrogen fertilizer is calculated by formula (4) and (5):

[0136] R=U×Y×ΔC (4)

[0137] ΔC=1-(C-CCV) / CCV (5)

[0138] Where: Y is the target yield; the amount of phosphorus and potassium is determined by the recommended fertilization ratio of nitrogen, phosphorus and potassium (1:0.8:1.5 during the fruit expansion period) and their respective ΔC.

[0139] Due to the lack of nutrient utilization rates in different orange orchards, this embodiment calculates three recommended fertilization treatments (RF1, RF2, and RF3) according to the average yield of low-yield orange orchards, the average yield of production areas, and the average yield of high-yield orange orchards in the data set. After verification, the treatment with the best comprehensive fruit quality is used as the main recommendation algorithm for the orange orchard.

[0140] 2.1.5 Compound fertilizer formula recommendation algorithm

[0141] In the early stage of the team's development of the agricultural supply company service platform, the compound fertilizer formula algorithm was used to generate a recommended fertilization plan for compound fertilizer based on the effective nutrient content of different types of fertilizers in the local fertilizer warehouse for macroelements, and the corresponding foliar spraying was used for correction based on the nutritional diagnosis results of the orange orchard for medium and trace elements.

[0142] 2.2 System Development

[0143] 2.2.1 Overall system structure

[0144] The development platform is Visual Studio Code and idea. The system website is built using spring+spring, boot+spring and cloud+flutter+vue+python frameworks. The overall structure of the citrus nutrition and fertilization recommendation expert system is as follows: Figure 4 The raw data of the orange orchard can be uploaded by entrusting a professional machine testing agency or providing government agencies with access to the corresponding database, and then transferred to the system server; based on Java1.8 / python 3.7.4, it can determine the corrected effective nutrient prescription recommendation plan according to the nutritional diagnosis and recommended fertilization decision process; and automatically connect to the agricultural material company service system through http / https to obtain the recommended fertilization decision plan based on the information released by local agricultural material merchants (such as Figure 5 The system database management adopts Mysql8.0 relational database; growers use B / S (Brower / Server) mode, connect to the system server through a web browser, and the interface is implemented by mixed programming of Html+JavaScript and flutter language.

[0145] 2.2.2 User Permission Settings

[0146] The platform uses role-based access control (RBAC) to set different roles to achieve different user rights allocation, as shown in Table 1. The backend system includes the update and upload of basic information and data information of the production area, and configures permissions for planting users, professional users and system administrators; among them, the permission to upload the sample test results entrusted by planting users is open to professional institutions, and the management permission of the system to connect to the production area database interface is open to the government agricultural technology extension department. The front-end system mainly displays the fertilization recommendation processing results of the expert system. Planting users can query the nutrient status of their own orange orchards, the recommended effective nutrient formula for fertilization, and the recommended fertilization plan for compound fertilizers based on the local agricultural materials library according to the location information; government agency users can view the mineral nutrition suitability standards, nutrient balance status evaluation and compound fertilizer recommendation plan summary of the orange orchards in their respective production areas.

[0147] Table 1 Role permissions table

[0148]

[0149] 3. System implementation and application

[0150] 3.1 Establishment of suitable standards for high-quality leaf nutrients in production areas

[0151] The growers or professional institutions upload the nutrient and fruit quality indicators of the fruit expansion period of Longan pomelo orchard in Guang'an District, Sichuan Province for two consecutive years from 2019 to 2020 as the original data to the system server (such as Figure 6 and Figure 7 ), the system calls the fruit quality comprehensive evaluation algorithm and the leaf mineral nutrition suitable standard generation algorithm in the algorithm model library to automatically generate the leaf nutrient suitable standard for the fruit expansion period as N 22.7~24.8g·kg -1 、P 1.1~1.3g·kg -1 、K9.9~13.3g·kg -1 , Ca 61.2~71.5g·kg -1 Mg 0.22~0.27g·kg -1 、Mn 35.1~85.8mg·kg -1 、Fe 4.1~7.8mg·kg -1 、Cu5.6~7.5mg·kg -1 、Zn 12.3~19.5g·kg -1 ,like Figure 8 As shown, the system front-end opens raw data viewing permissions to professional users and government agencies.

[0152] 3.2 Evaluation of nutritional balance status of orange orchards in production areas

[0153] Based on the 2021 Longan pomelo orchard inspection data of Longan Township and Chongwang Township, the suitable nutrient standards for high-quality leaves during the fruit expansion period constructed by this system are called, while the production areas that have not constructed suitable nutrient standards for leaves of different varieties during the fruit expansion period adopt unified standards, and apply the nutritional status diagnosis algorithm model of the orange orchard in this system. Professional users and government agencies are authorized to view the nutritional diagnosis index and results (such as Fig. 9As shown). The results showed that during the fruit expansion period of 2021, the nutrient excess limiting factors of the orange orchard in Long'an Township were K and Mn, with nutritional indexes of 9.45 and 6.96; the deficiency limiting factors were P, Ca and Zn elements, with nutritional indexes of -9.93, -8.64 and -6.40, respectively, and the fertilizer requirement order of the orange orchard was P>Ca>Zn. In the orange orchard (OF) in Chongwang Township, the nutrient excess limiting factor was P, with a nutritional index of 76.38, and the Mg element was high, with a nutritional index of 27.57; the deficiency limiting factors were Mn and Zn, with nutritional indexes of -54.73 and -42.48, respectively, and the N and K elements were low, with nutritional indexes of -21.82% and 9.30%, and the fertilizer requirement order was Mn>Zn>>N>K.

[0154] 3.3 Recommended effective nutrient formula and compound fertilizer scheme for orange orchards in production areas

[0155] According to the production statistics of Longan pomelo production area from 2019 to 2020, the low-yield garden, average yield and high-yield garden were 28.98, 35.71 and 47.78 t·hm respectively. -1 . Calling the nutritional diagnosis index of orange orchards in the production area, and applying the effective nutrient formula correction algorithm model of this system, the three effective nutrient correction formula schemes for the orange orchards in Long'an Township reduced the potassium input by 9.45% and increased the phosphorus input by 9.93% on the basis of the nutrient requirements for their respective target yields; the three effective nutrient correction formula schemes for the orange orchards in Chongwang Township reduced the phosphorus input by 76.38% and increased the nitrogen input by 21.82% and the potassium input by 9.30% on the basis of the nutrient requirements for their respective target yields. Calling the results of the compound fertilizer formula recommendation algorithm of the agricultural materials service system, planting users and government agencies are authorized to view the three effective nutrient correction formula schemes for orange orchards and the recommended compound fertilizer schemes based on the agricultural materials database provided by a certain company. This system only provides recommendations for macro-element fertilization for planting users and government agencies, and the fertilization of medium and trace elements is implemented according to the standards of the local agricultural technology extension department. The recommended page for compound fertilizer schemes is as follows: Fig.10 shown.

[0156] 3.4 Verification of the application effect of the expert system for recommending nutrient fertilization for citrus

[0157] According to the effective nutrient correction plan of the fertilization recommendation system in 2021, the effective nutrient application amount of each treatment in Long'an production area and Chongwang production area is shown in Table 2, and the appearance quality of Long'an pomelo under different fertilization treatments is shown in Table 3. The results show that the recommended fertilization treatment in Long'an Township significantly reduces the amount of chemical fertilizers compared with the empirical fertilization (CFP) based on chemical fertilizers. Among them, the nitrogen input of CRF1, CRF2 and CRF3 treatments decreased by 53.9%, 41.7% and 19.9%, respectively, and the potassium input decreased by 55.4%, 44.7% and 24.8%, respectively, and the phosphorus input of CRF1 and CRF2 decreased by 25.1% and 5.1%, respectively. In Chongwang Township, which used empirical fertilization (OFP) mainly based on organic fertilizers, the DOP indexes of nitrogen and potassium in the orange orchard were -21.82 and -9.3, respectively, and the DOP index of phosphorus was 76.38 ( Figure 5 ), so compared with the OFP treatment, the nitrogen input of the recommended fertilization treatments of ORF1, ORF2 and ORF3 increased by 54.1%, 93.8% and 160.1%, respectively, and the potassium input increased by 29.7%, 63.3% and 119.4%, respectively, to meet the nutrients for fruit growth and development and the correction requirements of N and K elements; but the phosphorus input decreased by 79.6%, 74.4% and 65.6%, respectively; however, compared with CFP, the nitrogen input of ORF1, ORF2 and ORF3 treatments in Chongwang Township decreased by 43.5%, 28.6% and 1.94%, respectively, the potassium input decreased by 48.1%, 35.7% and 12.4%, respectively, and the phosphorus input decreased by 82.2%, 77.5% and 70.4%, respectively. On this basis of saving fertilizer, according to the citrus fruit quality evaluation index algorithm established by the author's team, the comprehensive quality evaluation model of Longan pomelo is: Fq = 0.28×FMI-0.23×CCI+0.19×PFW+0.14×ER+0.09×TSS+0.04×Vc (standardized value). The CRF2 treatment in Longan Township orange orchard (CF) where chemical fertilizers were mainly used had the best effect. Compared with the CFP treatment, the soluble solids of the fruit increased significantly by 5.2%, and the titratable acid decreased significantly by 13.1%, resulting in a significant increase of 12.1% in the solid-acid ratio of the fruit (e.g. Figure 11(A)-Figure 11(F) As shown, Figure 11(A)-Figure 11(F) As shown, the horizontal axis in Figure 11 represents the type of treatment. The lowercase letters a, b, c, and d have no practical meaning, indicating that the quality indicators of different treatments are significantly different at the 5% level. For example, in Figure 11 (A), the letter for ORF2 is a, the letter for CRF2 is ab, and the letters for CFP and OFP are both c, which means that the TSS of CRF2 and ORF2 are significantly higher than those of CFP and OFP, while CRF2 and ORF2 both contain the common letter a, indicating that there is no significant difference in TSS between CRF2 and ORF2. ), the comprehensive quality index of the fruit is in the range of 0.55-0.71 (mainly first-class fruit); while the comprehensive quality index of the fruit treated with CFP is 0.19-0.44 (mainly third-class fruit) ( Figure 7The ORF2 treatment had the best effect in the orange orchard (OF) of Chongwang Township, where organic fertilizer was mainly used. Compared with OFP, the fruit soluble solids and fruit solid-acid ratio increased significantly by 9.3% and 12.4%, respectively (Appendix Figure 5 ), the comprehensive quality index of the fruit was in the range of 0.55-0.74 (mainly first-class fruit); while the comprehensive quality index of the fruit treated with OFP was in the range of 0.46-0.66 (mainly second-class fruit) (e.g. Fig.12 In addition, the appearance color index of the fruit in OFP treatment was significantly better than that in other treatments, resulting in its appearance coloration and fruit comprehensive quality index being significantly better than that in CFP treatment, but its single fruit weight was significantly less than that in other treatments, and the coloring of the flesh was hindered (e.g. Figure 8 , as shown in Table 2 and Table 3).

[0158] Table 2 Comparison of nutrient dosage in different treatments

[0159]

[0160] Table 3 Comparison of the appearance quality of Longan pomelo under different fertilization treatments

[0161]

[0162] 4 Discussion

[0163] China's citrus planting area and output have increased by 2.97 and 4.83 times in the past 20 years (2000-2020), accounting for 29.7% and 28.2% of the world's planting area and output, respectively. However, the rate of high-quality fruit is not high. With the change of residents' consumption concept, more attention is paid to fruit quality, which directly leads to the structural oversupply and undersupply of the citrus market.

[26] . In addition, the increase in the prices of production materials such as agricultural fertilizers has led to an increase in production costs, further squeezing the profit margins of fruit farmers and fruit dealers. Based on the current status of citrus nutrition diagnosis and fertilization, this study constructed an expert system for citrus nutrition fertilization recommendation, which aims to improve fruit quality, while taking into account the reduction of nutrients in production areas dominated by chemical fertilizers and the nutrient balance in production areas dominated by organic fertilizers. It has important application value in providing standardized fertilization planning for the same citrus production area, and is one of the important measures to achieve standardized fruit production and improve fruit competitiveness.

[0164] There are many crop precision fertilization algorithm models, but they are too complex for farmers, and the algorithm transplantation is difficult and costly. For example, Puntel uses machine learning to calculate the optimal amount of fertilizer for corn, but it is still in the experimental stage of professional institutions. This study established a citrus nutrition fertilization recommendation expert system based on the nutrition diagnosis formula fertilization algorithm established by the team in the early stage. The access of different users is controlled by role-based access control (RBAC). The original data is only granted to the uploaded user. Ordinary growers can directly view the three recommended fertilization plans of the system. Farmers adjust the recommended fertilization plan suitable for the local area according to the final verification results, which not only reduces the difficulty of transplanting the fertilization recommendation algorithm but also realizes the intellectual property protection of the data set and algorithm library.

[0165] The present invention uses the fruit comprehensive evaluation hierarchical analysis algorithm model based on principal component analysis established in the early stage to realize the comprehensive quality evaluation of orange orchards with different nutrient status in the modern agricultural production area of ​​Long'an pomelo in Guang'an City, and uses the inflection point value algorithm of the suitable standard of nutrients of orange orchard leaves to exclude orange orchards with poor quality caused by imbalance of nutrient levels, and constructs the suitable nutrient standard of high-quality leaves in the fruit expansion period applicable to the production area, combined with the "quantity determined by production, fertilizer determined by nitrogen" DOP effective nutrient correction algorithm, and finally realizes the recommendation of compound fertilizer scheme based on the local agricultural material warehouse, and authorizes the local agricultural material management department to view it, which is conducive to the dispatch of agricultural materials. At the same time, the application effect verification results of the citrus nutrition and fertilization recommendation expert system show that the region has a nutrient content of 35.71t·hm -1 The recommended fertilization scheme is the best with the average yield as the target (e.g. Fig.13 ), which can be used as the main recommended fertilization scheme, is significantly lower than the empirical fertilization input of chemical fertilizers (Table 2), achieving the purpose of "weight loss" in this study. In addition, it effectively avoids the delayed coloring of the peel and the low overall quality of the fruit caused by excessive and unbalanced fertilization input in the empirical orange orchards that mainly use chemical fertilizers; it also effectively avoids the low nitrogen and potassium fertilizer input in the empirical fertilization that mainly uses organic fertilizers, which leads to small fruits and delayed flesh coloring, and significantly improves the overall quality of the fruit in the Long'an pomelo production area.

[0166] 5 Conclusion

[0167] The present invention constructs an expert system for recommending citrus nutrition and fertilization, establishes an information service platform for the linkage of professional institutions, growers, agricultural supply companies and government agencies, realizes fertilization recommendations based on orchard nutrition diagnosis, and recommends treatments that have significant effects on reducing weight and improving quality in the modern agricultural production areas of Longan pomelo; among them, the recommended fertilization scheme based on the average yield of the production area has the best comprehensive fruit quality, and will be used as the main fertilization decision for the production area in the future. However, the current system only establishes local high-quality leaf nutrient suitability standards, nutrient balance status evaluation of orange orchard trees, and recommended fertilization schemes for the fruit expansion period. In the future, it is necessary to establish algorithm models such as high-yield leaf nutrient suitability standards and recommended fertilization decisions during the physiological fruit drop period to meet the fertilization needs of the orange orchard throughout the year.

[0168] Example 2

[0169] In order to execute the method corresponding to the above embodiment 1 to achieve corresponding functions and technical effects, a fertilization recommendation system is provided below, and the system includes:

[0170] The leaf content acquisition module is used to obtain the leaf content of multiple nutrient elements in the target orchard during the fruit growth period.

[0171] The standard leaf content retrieval module is used to retrieve the standard leaf content of multiple nutrient elements in the target orchard from the database.

[0172] The fertilization recommendation scheme determination module is used to determine the fertilization recommendation scheme of the target orchard based on multiple standard leaf contents and multiple leaf contents.

[0173] Example 3

[0174] This embodiment provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute a fertilization recommendation method described in Embodiment 1. The memory is a readable storage medium.

[0175] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0176] This article uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only used to help understand the method and core ideas of the present invention. At the same time, for those skilled in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A fertilization recommendation method, characterized in that: The fertilization recommendation method is applied to a database-based fertilization recommendation platform, which includes a system server and multiple user terminals; The system server is connected to the plurality of user terminals respectively; The user terminal is used to obtain leaf contents of multiple nutrient elements required by fruit trees in multiple orchards, and fruit quality in multiple orchards; The system server is used to construct a database based on the leaf contents of multiple nutrient elements required by fruit trees in multiple orchards and the fruit quality of multiple orchards; the database is used to store the standard leaf contents of multiple nutrient elements required by fruit trees in multiple orchards; The user terminal is used to obtain a target orchard fertilization recommendation plan request; The system server is used to generate a fertilization recommendation plan for the target orchard according to the fertilization recommendation plan request of the target orchard; The user terminal is also used to display the recommended fertilization plan; The method comprises: Obtain leaf contents of multiple nutrients in the fruit Pengda period target orchard; The standard leaf contents of multiple nutrient elements in the target orchard are retrieved from the database; Determining a recommended fertilization plan for a target orchard based on a plurality of said standard leaf contents and a plurality of said leaf contents; According to the plurality of standard leaf contents and the plurality of leaf contents, a recommended fertilization plan for a target orchard is determined, comprising: Determine any nutrient element as the current nutrient element; According to the standard leaf content and leaf content of the current nutrient elements, determine whether the current nutrient elements are the nutrient elements that need to be applied to the target orchard; When the current nutrient element is the target orchard that needs to apply the nutrient element, the current correction coefficient is determined using the formula ΔC=1-(C-CCV) / CCV according to the standard leaf content and leaf content of the current nutrient element; wherein ΔC is the current correction coefficient; C is the leaf content of the current nutrient element; and CCV is the standard leaf content of the current nutrient element. According to the current correction coefficient, the formula R=U×Y×ΔC is used to determine the fertilizer requirement of the current nutrient element; U is the nutrient absorption of the fruit tree for the current nutrient element during the fruit growth period; and Y is the target fruit yield. Update the current nutrient element, and return to step "determine whether the current nutrient element is a nutrient element that needs to be applied to the target orchard based on the standard leaf content and leaf content of the current nutrient element"; Determine the recommended fertilization plan for the target orchard based on the nutrients to be applied and the required amount of nutrients to be applied.

2. A fertilization recommendation method according to claim 1, characterized in that: The standard leaf contents of multiple nutrient elements of the target orchard are retrieved from the database, including: Determine whether the database stores standard leaf contents of multiple nutrient elements required by fruit trees in the target orchard, and obtain a first determination result; If the first judgment result is yes, then the standard leaf contents of multiple nutrient elements of the target orchard are retrieved from the database; If the first judgment result is no, determining the orchard closest to the target orchard as the target replacement orchard; The standard leaf contents of various nutrients in the target replacement orchard are retrieved as the standard leaf contents of various nutrients in the target orchard.

3. A fertilization recommendation method according to claim 1, characterized in that: Before obtaining the leaf content of multiple nutrients in the fruit Peng Daqi target orchard, it also includes: The database is constructed.

4. A fertilization recommendation method according to claim 3, characterized in that: The constructing of the database comprises: Determine any orchard as the current orchard; Get the current orchard fruit quality; Obtain the leaf content of multiple nutrients in the current orchard of Fruit Peng Daqi; According to the current orchard fruit quality and leaf contents of multiple nutrients, the nutrient component analysis method was used to determine the proposed standard leaf content of each nutrient element.

5. A fertilization recommendation method according to claim 4, characterized in that: The obtaining of the current orchard fruit quality includes: Obtain multiple fruit quality indicators of multiple sample fruits in the current orchard during the fruit maturity period; Determine the sum of the products of each fruit quality index and the corresponding weight as the sample fruit quality; The average value of the sample fruit quality was determined as the current orchard fruit quality.

6. A fertilization recommendation method according to claim 1, characterized in that: The step of determining whether the current nutrient element is a nutrient element that needs to be applied to the target orchard according to the standard leaf content and leaf content of the current nutrient element comprises: Determine whether the absolute value of the difference between the standard leaf content and the leaf content of the current nutrient element is less than a preset difference threshold, and obtain a second determination result; If the second judgment result is yes, then determine that the current nutrient element does not need to be added; update the current nutrient element, If the second judgment result is no, it is determined that the current nutrient element is a nutrient element that needs to be added.

7. A fertilization recommendation system, characterized in that: The fertilization recommendation system applies the fertilization recommendation method according to any one of claims 1 to 6, and the fertilization recommendation system includes: Leaf content acquisition module, used to obtain the leaf content of multiple nutrient elements in the target orchard during the fruit growth period; A standard leaf content retrieval module is used to retrieve the standard leaf content of multiple nutrient elements in the target orchard from the database; The fertilization recommendation scheme determination module is used to determine the fertilization recommendation scheme of the target orchard according to the multiple standard leaf contents and the multiple leaf contents; the fertilization recommendation scheme determination module is also used to determine any nutrient element as the current nutrient element; determine whether the current nutrient element is the nutrient element that needs to be applied to the target orchard according to the standard leaf content and the leaf content of the current nutrient element; when the current nutrient element is the nutrient element that needs to be applied to the target orchard, determine the current correction coefficient according to the standard leaf content and the leaf content of the current nutrient element using the formula ΔC=1-(C-CCV) / CCV; wherein ΔC is the current correction coefficient; C is The current leaf content of the nutrient element; CCV is the standard leaf content of the current nutrient element; according to the current correction coefficient, the fertilizer requirement of the current nutrient element is determined using the formula R=U×Y×ΔC; wherein R is the fertilizer requirement of the current nutrient element; U is the nutrient absorption of the current nutrient element by the fruit tree during the fruit expansion period; Y is the target fruit yield; update the current nutrient element, and return to the step "determine whether the current nutrient element is the nutrient element that needs to be applied to the target orchard based on the standard leaf content and leaf content of the current nutrient element"; determine the recommended fertilization plan for the target orchard based on the nutrient element that needs to be applied and the fertilizer requirement of the nutrient element that needs to be applied.

8. An electronic device, characterized in that: The electronic device comprises a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute a fertilization recommendation method according to any one of claims 1 to 6.

9. An electronic device according to claim 8, characterized in that: The memory is a readable storage medium.

Images