Agricultural desertification control method for apple planting area

Abstract

The invention discloses an apple planting area agricultural desertification control method, and relates to the technical field of desertification control, and the method comprises the following steps: S1, collecting soil data, S2, calculating development parameters, S3, determining genetic parameters, S4, analyzing variety parameters, S5, optimizing environmental conditions, and S6, monitoring crop growth. According to the method, three different models are used for evaluation, the optimal temperature difference, the optimal total daily radiation, the shortest growth index standard reaching date and other conditions can be rapidly determined, an optimization thought is provided for apple tree planting according to the data, measures are taken for intervention, the soil environment of apple planting is improved, and the apple planting quality is improved. According to the method, the apple planting yield is increased to a certain extent, meanwhile, the apples are monitored in the apple growth process, monitored data are transmitted to a visual interface and a database, an operator can conveniently know the actual planting condition in time, and meanwhile data support is provided for subsequent model analysis.

Description

Technical Field

[0001] The invention relates to the technical field of desertification control, and in particular to a method for controlling agricultural desertification in an apple-planting area. Background Art

[0002] The process of growing sand apples in the desert involves efforts and practices in many aspects, including soil and water conservation, vegetation restoration, and agricultural technology improvement, etc., to improve the original soil conditions so that it can support plant growth. In the invention patent with application number CN202011098143.2, it is disclosed that "a method for controlling land desertification includes water source collection: collecting fresh water and irrigable sewage separately; water source transportation: transporting the collected fresh water and irrigable sewage to the desertified areas that need to be controlled; water source storage: storing fresh water in an underground intelligent irrigation structure established in the desertified areas. , the irrigable sewage is stored in the artificial sewage lake on the ground; irrigation: the underground intelligent irrigation structure will store the fresh water in it through the outlet on the underground water pipe for intelligent irrigation, and the sewage in the artificial sewage lake on the ground will be sprayed for irrigation through the drainage structure connected to the underground pipeline. The present invention utilizes seawater desalination technology, weather forecast system and sewage treatment technology to utilize seawater, floods and urban domestic sewage, which can not only solve the problem of land desertification, but also make full use of various water resources, reduce the rise of seawater level, the loss of life and property caused by floods, and the pollution caused by sewage discharge. "

[0003] The above-mentioned existing technology solves the problems of serious desertification and degradation of limited arable land and natural grasslands. However, in actual application, this method is relatively simple in that it only uses water sources to improve soil conditions. It is unable to obtain economic benefits after treatment. Moreover, before apple planting, it is impossible to simulate the growth process of apples according to the current soil conditions, resulting in the operator being unable to accurately know the growth cycle of apples, affecting the final planting effect. At the same time, it is also impossible to determine the optimal environmental conditions for apple planting, reducing the yield of apple planting. Summary of the invention

[0004] The purpose of the present invention is to provide a method for controlling agricultural desertification in apple-growing areas to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a method for controlling agricultural desertification in apple planting areas, comprising the following steps:

[0006] S1. Collect soil data: transfer daily meteorological data and corresponding dates to a storage device, use a soil analyzer to test the soil at different depths in each sub-area, obtain the corresponding soil parameters and component proportions, select the current apple tree planting variety and soil parameters and component proportions, extract the corresponding growth index from the historical database, and determine the seedling stage, flowering stage, maturity stage and yield of the variety;

[0007] S2. Calculate growth parameters: After the apple variety and related data are transmitted to the apple basic growth model, the initial parameters corresponding to the variety and the growth index in the time period are output after analysis, and the actual growth index corresponding to the same time period is transmitted to the apple basic growth model as correction data, so as to deduce the optimal growth parameters under the current soil parameters and component proportions;

[0008] S3. Determine genetic parameters: Analyze apple varieties and related data using the apple optimization model, output the emergence period, flowering period, maturity period and yield of the crop, calculate the genetic parameters corresponding to the variety based on the simulation data and actual data, re-simulate using the genetic parameters, calculate the actual value and the output result through the indicator analysis algorithm, and if all the indicators obtained are greater than the preset values, determine that the current parameters are the optimal genetic parameters;

[0009] S4. Analyze variety parameters: determine the type and range of each variety parameter, select multiple values ​​within the corresponding range of each parameter using random sampling method, and after assigning values ​​to the parameters, analyze the input data through the apple analysis model, transfer the results to the database, and use the likelihood equation to calculate the deviation between the model output result and the actual growth index. Repeat the operation until all values ​​in the set are selected, and take the parameter combination corresponding to the output result with the smallest deviation value as the final variety parameter;

[0010] S5. Optimize environmental conditions: After obtaining the condition parameters to be optimized, initialize the corresponding particles according to the parameters, determine the corresponding model interface by the parameter type, obtain the fitness corresponding to the current particle after model evaluation, calculate the new speed and position corresponding to the current particle using the particle optimization algorithm, re-evaluate, and calculate the new fitness. If the new fitness is lower than the current fitness and the current fitness is higher than the threshold, determine the optimal particle speed and position, and output the name, type and optimal value of the corresponding condition parameter according to the particle information;

[0011] S6. Monitoring crop growth: Extract the original spectrum, vegetation index, spatial texture and color space feature set corresponding to the CF-6WFV image data of each sub-area. After analysis by the apple recognition model, output the apple growth index corresponding to the current image, and transmit the sub-area number, apple growth index and current date to the database.

[0012] Preferably, the S1 specifically comprises the following steps:

[0013] S101, after querying the daily meteorological data of the specified area in the meteorological data platform, the data and the corresponding date are transmitted to the storage device, wherein the daily meteorological data includes the maximum temperature, the minimum temperature, the precipitation and the sunshine time throughout the day;

[0014] S102, dividing the designated area into multiple sub-areas, numbering each sub-area, setting four sampling depths, namely 6 cm, 18 cm, 30 cm and 42 cm, collecting soil at different depths of the sub-areas in sequence according to the numbers, testing the soil using a soil analyzer, and obtaining soil parameters and component proportions corresponding to different depths of each sub-area;

[0015] S103. After collecting the soil characteristic data set using the crawler tool, transfer it to the storage device, select the planting variety, soil parameters and component ratios of the apple trees, and extract the corresponding growth index in each time period under the soil parameters and component ratios of the variety from the historical database to determine the emergence period, flowering period, maturity period and yield of the variety.

[0016] Preferably, the S2 specifically includes the following steps:

[0017] S201, after transmitting the apple variety, current soil parameters and component proportions and daily weather data to the apple basic growth model, setting the start time and end time of the simulation, the model outputs the initial parameters corresponding to the variety and the growth index within the time period after analysis, and the apple basic growth model is specifically the WOFOST model;

[0018] S202. Determine the actual growth index based on the current soil parameters and component proportions and the corresponding time period of the simulation, transmit these data as correction data to the apple basic growth model, and use the output growth index and the actual growth index to calculate the optimal development parameters under the current soil parameters and component proportions.

[0019] Preferably, the S3 specifically includes the following steps:

[0020] S301, calculating the total radiation value according to the full-day sunshine time in the daily meteorological data, and transmitting the soil parameters and component proportions, daily meteorological data, total radiation value and apple varieties to the apple optimization model, wherein the apple optimization model is specifically an APSIM model;

[0021] S302, after model analysis, the emergence period, flowering period, maturity period and yield of the crop are output, and the emergence period, flowering period, maturity period and yield of the actual crop are obtained according to the current soil parameters and component proportions and the specific date corresponding to the daily meteorological data, and the genetic parameters corresponding to the variety are calculated according to the simulated data and the actual data;

[0022] S303, reanalyze the emergence period, flowering period, maturity period and yield of the crop according to the genetic parameters through the model, calculate the actual value and the simulated value using the index analysis algorithm, and if all the obtained indicators are greater than the preset values, determine that the current parameters are the optimal genetic parameters, otherwise recalculate the genetic parameters, and the index analysis algorithm is specifically as follows:

[0023]

[0024] Among them, Y i Represents the analog value, X i Indicates the measured value, represents the average value of the measured data, n represents the total number of data, i represents the parameter, R represents the relative error index, N represents the absolute error index, and D represents the consistency index.

[0025] Preferably, the S4 specifically comprises the following steps:

[0026] S401, after transmitting soil parameters and component proportions, daily meteorological data, total radiation values ​​and crop varieties to the apple analysis model, determine the type and range of each variety parameter, select multiple values ​​within the corresponding range of each parameter using a random sampling method, construct a corresponding set for each parameter, and store these values ​​in the set, wherein the apple analysis model is specifically a DSSAT model;

[0027] S402, selecting values ​​from the set, assigning values ​​to the parameters, using the model to analyze the input soil parameters and component proportions, daily meteorological data, total radiation values ​​and crop varieties, and transmitting the output results to the database;

[0028] S403, according to the current soil parameters, component proportions and specific dates, determine the corresponding actual growth index, use the likelihood equation to calculate the deviation between the model output result and the actual growth index, repeat the operation until all values ​​in the set are selected, count the deviation values ​​corresponding to all parameter combinations, and use the parameter combination corresponding to the output result with the smallest deviation value as the final variety parameter;

[0029] S404. After adding the proportion of fertilizer application to the model, the growth index of crops under different proportions of fertilizer application is simulated using the apple analysis model. The output results are plotted into a curve graph and then output through a visual interface.

[0030] Preferably, the S5 specifically includes the following steps:

[0031] S501, after obtaining the name, type, range and number of condition parameters to be optimized, each particle is initialized according to these data, and the corresponding model interface is determined according to the parameter type;

[0032] S502. If the parameter type is associated with the developmental parameter type, the basic apple growth model is used to evaluate the numerical value corresponding to the particle. If the parameter type is associated with the genetic parameter type, the apple optimization model is used to evaluate it. If the parameter type is associated with the variety parameter, the apple analysis model is used to evaluate it.

[0033] Preferably, the S5 further comprises the following steps:

[0034] S503, after evaluation, the fitness corresponding to the current particle is obtained, the fitness is used as the initial fitness, and the particle optimization algorithm is used to calculate according to the speed and position of the current particle to obtain the new speed and position corresponding to the current particle, and the model is used to re-evaluate to obtain a new fitness;

[0035] S504. If the new fitness is higher than the current fitness, the current fitness and the speed and position of the particle are updated. If the new fitness is lower than the current fitness and the current fitness is higher than the threshold, the optimal particle speed and position are determined, and the name, type and optimal value of the corresponding conditional parameter are output according to the particle information, speed and position.

[0036] Preferably, the S6 specifically comprises the following steps:

[0037] S601, obtaining CF-6WFV image data corresponding to each sub-region, setting a threshold, removing pixels corresponding to non-crops contained in the image data, and extracting the original spectrum, vegetation index, spatial texture and color space feature set corresponding to the image data;

[0038] S602, constructing an apple recognition model using BP neural network technology, transmitting a feature set corresponding to the CF-6WFV image data to the model, and outputting an apple growth index corresponding to the current image after model recognition;

[0039] S603, transmitting the sub-region number, apple growth index and current date to the database, calculating the apple growth index of all regions within the current date, and outputting it through a visual interface;

[0040] S604: Draw a corresponding curve graph according to the crop growth index predicted by the apple basic growth model and the actual growth index of each sub-region, and output the curve graph through a visualization interface.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] 1. The present invention samples soil at different depths. Due to the increase in the number of soil sampling, a more accurate growth index can be matched when selecting soil parameters and component proportions. The historical data of the same variety, soil parameters and component proportions can be used to invert and deduce the optimal developmental parameters and genetic parameters in the current apple basic growth model and apple optimization model, so that the subsequent models can simulate the growth environment with the same soil parameters and component proportions more accurately and fit the actual growth conditions of apple trees. At the same time, the variety parameters in the apple analysis model are obtained by calculating the deviation between the model output result and the actual growth index using the likelihood equation, and the parameter combination corresponding to the output result with the smallest deviation value is used as the final variety parameter. This method can simultaneously determine the values ​​of multiple parameters of the model without requiring too much historical data, thereby avoiding the selected parameters being unsuitable for simulation of the current soil environment due to lack of data.

[0043] 2. The present invention utilizes three different types of models for evaluation, which can relatively quickly determine the optimal temperature difference, the optimal total daily radiation, the shortest date for reaching the growth index standard, and other conditions. Based on these data, optimization ideas are provided for the planting of apple trees, and intervention measures are taken to improve the soil environment for apple planting, thereby increasing the yield of apple planting to a certain extent. At the same time, apples are monitored during their growth process, and the monitored data is transmitted to a visualization interface and a database, so that operators can understand the actual planting situation in a timely manner, and data support is provided for subsequent model analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 Provide an overall method flow chart for an embodiment of the present invention;

[0045] Figure 2 A flow chart of a method for collecting soil data provided by an embodiment of the present invention;

[0046] Figure 3 A flow chart of a method for determining genetic parameters provided by an embodiment of the present invention;

[0047] Figure 4 A flow chart of a method for analyzing variety parameters provided by an embodiment of the present invention;

[0048] Figure 5 A flow chart of a method for optimizing environmental conditions provided by an embodiment of the present invention;

[0049] Figure 6 A flow chart of a method for monitoring crop growth provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0050] 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.

[0051] See also Figure 1-6 The present invention provides a technical solution: a method for controlling agricultural desertification in apple planting areas, comprising the following steps:

[0052] S1. Collect soil data: transfer daily meteorological data and corresponding dates to a storage device, use a soil analyzer to test the soil at different depths in each sub-area, obtain the corresponding soil parameters and component proportions, select the current apple tree planting variety and soil parameters and component proportions, extract the corresponding growth index from the historical database, and determine the seedling stage, flowering stage, maturity stage and yield of the variety;

[0053] S2. Calculate growth parameters: After the apple variety and related data are transmitted to the apple basic growth model, the initial parameters corresponding to the variety and the growth index in the time period are output after analysis, and the actual growth index corresponding to the same time period is transmitted to the apple basic growth model as correction data, so as to deduce the optimal growth parameters under the current soil parameters and component proportions;

[0054] S3. Determine genetic parameters: Analyze apple varieties and related data using the apple optimization model, output the emergence period, flowering period, maturity period and yield of the crop, calculate the genetic parameters corresponding to the variety based on the simulation data and actual data, re-simulate using the genetic parameters, calculate the actual value and the output result through the indicator analysis algorithm, and if all the indicators obtained are greater than the preset values, determine that the current parameters are the optimal genetic parameters;

[0055] S4. Analyze variety parameters: determine the type and range of each variety parameter, select multiple values ​​within the corresponding range of each parameter using random sampling method, and after assigning values ​​to the parameters, analyze the input data through the apple analysis model, transfer the results to the database, and use the likelihood equation to calculate the deviation between the model output result and the actual growth index. Repeat the operation until all values ​​in the set are selected, and take the parameter combination corresponding to the output result with the smallest deviation value as the final variety parameter;

[0056] S5. Optimize environmental conditions: After obtaining the condition parameters to be optimized, initialize the corresponding particles according to the parameters, determine the corresponding model interface by the parameter type, obtain the fitness corresponding to the current particle after model evaluation, calculate the new speed and position corresponding to the current particle using the particle optimization algorithm, re-evaluate, and calculate the new fitness. If the new fitness is lower than the current fitness and the current fitness is higher than the threshold, determine the optimal particle speed and position, and output the name, type and optimal value of the corresponding condition parameter according to the particle information;

[0057] S6. Monitoring crop growth: Extract the original spectrum, vegetation index, spatial texture and color space feature set corresponding to the CF-6WFV image data of each sub-area. After analysis by the apple recognition model, output the apple growth index corresponding to the current image, and transmit the sub-area number, apple growth index and current date to the database.

[0058] S1 specifically includes the following steps:

[0059] S101, after querying the daily meteorological data of the specified area in the meteorological data platform, the data and the corresponding date are transmitted to the storage device, wherein the daily meteorological data includes the maximum temperature, the minimum temperature, the precipitation and the sunshine time throughout the day;

[0060] S102, dividing the designated area into multiple sub-areas, numbering each sub-area, setting four sampling depths, namely 6 cm, 18 cm, 30 cm and 42 cm, collecting soil at different depths of the sub-areas in sequence according to the numbers, testing the soil using a soil analyzer, and obtaining soil parameters and component proportions corresponding to different depths of each sub-area;

[0061] S103, after collecting the soil property data set using the crawler tool, transfer it to a storage device, select the planting variety and soil parameters and component proportions of the apple tree, extract the soil parameters and component proportions of the variety in the historical database according to the growth index corresponding to each time period, and determine the seedling period, flowering period, maturity period and yield of the variety;

[0062] S2 specifically includes the following steps:

[0063] S201, after transmitting the apple variety, current soil parameters and component proportions, and daily weather data to the apple basic growth model, set the start time and end time of the simulation, and after analysis, the model outputs the initial parameters corresponding to the variety and the growth index within the time period. The apple basic growth model is specifically the WOFOST model;

[0064] S202, determining the actual growth index according to the current soil parameters and component proportions and the simulation corresponding time period, transmitting these data as correction data to the apple basic growth model, and calculating the optimal development parameters under the current soil parameters and component proportions using the output growth index and the actual growth index;

[0065] S3 specifically includes the following steps:

[0066] S301, calculating the total radiation value according to the full-day sunshine time in the daily meteorological data, and transmitting the soil parameters and component proportions, daily meteorological data, total radiation value and apple varieties to the apple optimization model, which is specifically the APSIM model;

[0067] S302, after model analysis, the emergence period, flowering period, maturity period and yield of the crop are output, and the emergence period, flowering period, maturity period and yield of the actual crop are obtained according to the current soil parameters and component proportions and the specific date corresponding to the daily meteorological data, and the genetic parameters corresponding to the variety are calculated according to the simulated data and the actual data;

[0068] S303, reanalyze the emergence period, flowering period, maturity period and yield of the crop according to the genetic parameters through the model, and calculate the actual value and the simulated value using the index analysis algorithm. If all the obtained indicators are greater than the preset values, the current parameters are determined to be the optimal genetic parameters, otherwise the genetic parameters are recalculated. The index analysis algorithm is specifically as follows:

[0069]

[0070] Among them, Y i Represents the analog value, X i Indicates the measured value, represents the average value of measured data, n represents the total number of data, i represents the parameter, R represents the relative error index, N represents the absolute error index, and D represents the consistency index;

[0071] S4 specifically includes the following steps:

[0072] S401, after transmitting soil parameters and component proportions, daily meteorological data, total radiation values ​​and crop varieties to the apple analysis model, determine the type and range of each variety parameter, select multiple values ​​within the corresponding range of each parameter using a random sampling method, construct a corresponding set for each parameter, and store these values ​​in the set. The apple analysis model is specifically a DSSAT model;

[0073] S402, selecting values ​​from the set, assigning values ​​to the parameters, using the model to analyze the input soil parameters and component proportions, daily meteorological data, total radiation values ​​and crop varieties, and transmitting the output results to the database;

[0074] S403, according to the current soil parameters, component proportions and specific dates, determine the corresponding actual growth index, use the likelihood equation to calculate the deviation between the model output result and the actual growth index, repeat the operation until all values ​​in the set are selected, count the deviation values ​​corresponding to all parameter combinations, and use the parameter combination corresponding to the output result with the smallest deviation value as the final variety parameter;

[0075] S404, after adding the proportion of fertilizer application to the model, the growth index of crops under different proportions of fertilizer application is simulated using the apple analysis model, and the output results are plotted into a curve graph and output through a visual interface;

[0076] S5 specifically includes the following steps:

[0077] S501, after obtaining the name, type, range and number of condition parameters to be optimized, each particle is initialized according to these data, and the corresponding model interface is determined according to the parameter type;

[0078] S502. If the parameter type is associated with the development parameter type, the value corresponding to the particle is evaluated using the apple basic growth model; if the parameter type is associated with the genetic parameter type, the apple optimization model is used to evaluate it; if the parameter type is associated with the variety parameter, the apple analysis model is used to evaluate it;

[0079] S5 specifically further includes the following steps:

[0080] S503, after evaluation, the fitness corresponding to the current particle is obtained, and the fitness is used as the initial fitness. The particle optimization algorithm is used to calculate according to the speed and position of the current particle to obtain the new speed and position corresponding to the current particle, and the model is used to re-evaluate to obtain a new fitness. The particle optimization algorithm is specifically as follows:

[0081] v i+1 =v i +c 1 *rand()*(pbest i -x i )+c 2 *rand()*(gbest i -x i )

[0082] x i+1 =v i +x i

[0083] Among them, v i Indicates the current particle speed, v i+1 Indicates the next speed of the current particle, xi Indicates the current particle position, x i+1 represents the next position of the current particle, c 1 and c 2 represents the learning factor, rand() represents the random generation function, pbest i and gbest i Indicates boundary value;

[0084] S504, if the new fitness is higher than the current fitness, then update the current fitness and the speed and position of the particle; if the new fitness is lower than the current fitness and the current fitness is higher than the threshold, then determine the optimal particle speed and position, and output the name, type and optimal value of the corresponding condition parameter according to the particle information, speed and position;

[0085] S6 specifically includes the following steps:

[0086] S601, obtaining CF-6WFV image data corresponding to each sub-region, setting a threshold, removing pixels corresponding to non-crops contained in the image data, and extracting the original spectrum, vegetation index, spatial texture and color space feature set corresponding to the image data;

[0087] S602, constructing an apple recognition model using BP neural network technology, transmitting a feature set corresponding to the CF-6WFV image data to the model, and outputting an apple growth index corresponding to the current image after model recognition;

[0088] S603, transmitting the sub-region number, apple growth index and current date to the database, calculating the apple growth index of all regions within the current date, and outputting it through a visual interface;

[0089] S604: Draw a corresponding curve graph according to the crop growth index predicted by the apple basic growth model and the actual growth index of each sub-region, and output the curve graph through a visualization interface.

[0090] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0091] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for controlling agricultural desertification in apple-growing areas, characterized in that: The method comprises the following steps: S1. Collect soil data: transfer daily meteorological data and corresponding dates to a storage device, use a soil analyzer to test the soil at different depths in each sub-area, obtain the corresponding soil parameters and component proportions, select the current apple tree planting variety and soil parameters and component proportions, extract the corresponding growth index from the historical database, and determine the seedling stage, flowering stage, maturity stage and yield of the variety; S2. Calculate growth parameters: After the apple variety and related data are transmitted to the apple basic growth model, the initial parameters corresponding to the variety and the growth index in the time period are output after analysis, and the actual growth index corresponding to the same time period is transmitted to the apple basic growth model as correction data, so as to deduce the optimal growth parameters under the current soil parameters and component proportions; S3. Determine genetic parameters: Analyze apple varieties and related data using the apple optimization model, output the emergence period, flowering period, maturity period and yield of the crop, calculate the genetic parameters corresponding to the variety based on the simulation data and actual data, re-simulate using the genetic parameters, calculate the actual value and the output result through the indicator analysis algorithm, and if all the indicators obtained are greater than the preset values, determine that the current parameters are the optimal genetic parameters; S4. Analyze variety parameters: determine the type and range of each variety parameter, select multiple values ​​within the corresponding range of each parameter using random sampling method, and after assigning values ​​to the parameters, analyze the input data through the apple analysis model, transfer the results to the database, and use the likelihood equation to calculate the deviation between the model output result and the actual growth index. Repeat the operation until all values ​​in the set are selected, and take the parameter combination corresponding to the output result with the smallest deviation value as the final variety parameter; S5. Optimize environmental conditions: After obtaining the condition parameters to be optimized, initialize the corresponding particles according to the parameters, determine the corresponding model interface by the parameter type, obtain the fitness corresponding to the current particle after model evaluation, calculate the new speed and position corresponding to the current particle using the particle optimization algorithm, re-evaluate, and calculate the new fitness. If the new fitness is lower than the current fitness and the current fitness is higher than the threshold, determine the optimal particle speed and position, and output the name, type and optimal value of the corresponding condition parameter according to the particle information; S6. Monitoring crop growth: Extract the original spectrum, vegetation index, spatial texture and color space feature set corresponding to the CF-6WFV image data of each sub-area. After analysis by the apple recognition model, output the apple growth index corresponding to the current image, and transmit the sub-area number, apple growth index and current date to the database.

2. The method for controlling agricultural desertification in apple-growing areas according to claim 1, characterized in that: The S1 specifically includes the following steps: S101, after querying the daily meteorological data of the specified area in the meteorological data platform, the data and the corresponding date are transmitted to the storage device, wherein the daily meteorological data includes the maximum temperature, the minimum temperature, the precipitation and the sunshine time throughout the day; S102, dividing the designated area into multiple sub-areas, numbering each sub-area, setting four sampling depths, namely 6 cm, 18 cm, 30 cm and 42 cm, collecting soil at different depths of the sub-areas in sequence according to the numbers, testing the soil using a soil analyzer, and obtaining soil parameters and component proportions corresponding to different depths of each sub-area; S103. After collecting the soil characteristic data set using the crawler tool, transfer it to the storage device, select the planting variety, soil parameters and component ratios of the apple trees, and extract the corresponding growth index in each time period under the soil parameters and component ratios of the variety from the historical database to determine the emergence period, flowering period, maturity period and yield of the variety.

3. The method for controlling agricultural desertification in apple-growing areas according to claim 1, characterized in that: The S2 specifically includes the following steps: S201, after transmitting the apple variety, current soil parameters and component proportions, and daily weather data to the apple basic growth model, set the start time and end time of the simulation, and after analysis, the model outputs the initial parameters corresponding to the variety and the growth index within the time period; S202. Determine the actual growth index based on the current soil parameters and component proportions and the corresponding time period of the simulation, transmit these data as correction data to the apple basic growth model, and use the output growth index and the actual growth index to calculate the optimal development parameters under the current soil parameters and component proportions.

4. The method for controlling agricultural desertification in apple-growing areas according to claim 1, characterized in that: The S3 specifically includes the following steps: S301, calculating the total radiation value according to the full-day sunshine time in the daily meteorological data, and transmitting the soil parameters and component proportions, daily meteorological data, total radiation value and apple varieties to the apple optimization model; S302, after model analysis, the emergence period, flowering period, maturity period and yield of the crop are output, and the emergence period, flowering period, maturity period and yield of the actual crop are obtained according to the current soil parameters and component proportions and the specific date corresponding to the daily meteorological data, and the genetic parameters corresponding to the variety are calculated according to the simulated data and the actual data; S303. Reanalyze the emergence period, flowering period, maturity period and yield of the crop according to the genetic parameters through the model, and calculate the actual value and the simulated value using the indicator analysis algorithm. If all the indicators obtained are greater than the preset values, the current parameters are determined to be the optimal genetic parameters, otherwise the genetic parameters are recalculated.

5. The method for controlling agricultural desertification in apple-growing areas according to claim 1, characterized in that: The S4 specifically comprises the following steps: S401, after transmitting soil parameters and component proportions, daily meteorological data, total radiation values ​​and crop varieties to the apple analysis model, determine the type and range of each variety parameter, select multiple values ​​within the corresponding range of each parameter using a random sampling method, construct a corresponding set for each parameter, and store these values ​​in the set; S402, selecting values ​​from the set, assigning values ​​to the parameters, using the model to analyze the input soil parameters and component proportions, daily meteorological data, total radiation values ​​and crop varieties, and transmitting the output results to the database; S403, according to the current soil parameters, component proportions and specific dates, determine the corresponding actual growth index, use the likelihood equation to calculate the deviation between the model output result and the actual growth index, repeat the operation until all values ​​in the set are selected, count the deviation values ​​corresponding to all parameter combinations, and use the parameter combination corresponding to the output result with the smallest deviation value as the final variety parameter; S404. After adding the proportion of fertilizer application to the model, the growth index of crops under different proportions of fertilizer application is simulated using the apple analysis model. The output results are plotted into a curve graph and then output through a visual interface.

6. The method for controlling agricultural desertification in apple-growing areas according to claim 1, characterized in that: The S5 specifically includes the following steps: S501, after obtaining the name, type, range and number of condition parameters to be optimized, each particle is initialized according to these data, and the corresponding model interface is determined according to the parameter type; S502. If the parameter type is associated with the developmental parameter type, the basic apple growth model is used to evaluate the numerical value corresponding to the particle. If the parameter type is associated with the genetic parameter type, the apple optimization model is used to evaluate it. If the parameter type is associated with the variety parameter, the apple analysis model is used to evaluate it.

7. The method for controlling agricultural desertification in apple-growing areas according to claim 6, characterized in that: The S5 specifically further comprises the following steps: S503, after evaluation, the fitness corresponding to the current particle is obtained, the fitness is used as the initial fitness, and the particle optimization algorithm is used to calculate according to the speed and position of the current particle to obtain the new speed and position corresponding to the current particle, and the model is used to re-evaluate to obtain a new fitness; S504. If the new fitness is higher than the current fitness, the current fitness and the speed and position of the particle are updated. If the new fitness is lower than the current fitness and the current fitness is higher than the threshold, the optimal particle speed and position are determined, and the name, type and optimal value of the corresponding conditional parameter are output according to the particle information, speed and position.

8. The method for controlling agricultural desertification in apple-growing areas according to claim 1, characterized in that: The S6 specifically comprises the following steps: S601, obtaining CF-6WFV image data corresponding to each sub-region, setting a threshold, removing pixels corresponding to non-crops contained in the image data, and extracting the original spectrum, vegetation index, spatial texture and color space feature set corresponding to the image data; S602, constructing an apple recognition model using BP neural network technology, transmitting a feature set corresponding to the CF-6WFV image data to the model, and outputting an apple growth index corresponding to the current image after model recognition; S603, transmitting the sub-region number, apple growth index and current date to the database, calculating the apple growth index of all regions within the current date, and outputting it through a visual interface; S604: Draw a corresponding curve graph according to the crop growth index predicted by the apple basic growth model and the actual growth index of each sub-region, and output the curve graph through a visualization interface.

Images