Application method and device of sustained and controlled release pesticide

Abstract

The invention relates to the technical field of novel pesticide application, in particular to an application method and device of a sustained and controlled release pesticide. Analyzing the inhibition degree of the pest species reproduction activity scale on the plant species growth rate through the application area chart, and estimating and analyzing the application dosage based on the inhibition degree to obtain the target application dosage of the sustained and controlled release pesticide required by each application area in the application area chart; constructing a current trajectory decision-making tree, predicting and fitting a propagation activity prediction area of insect pest species in the multi-dimensional space field by using the current trajectory decision-making tree, and screening application point locations of the propagation activity prediction area according to a sustained and controlled release action range of a sustained and controlled release pesticide preset monomer dosage to obtain accurate application point locations; and formulating a reasonable application time point of the sustained and controlled release pesticide through a hot spot distribution density-time frame kurtosis trend model. According to the invention, the application of the sustained and controlled release pesticide can be precisely and reasonably planned and optimized, so that the pesticide effect quality of the sustained and controlled release pesticide on insect pests is improved, and the sustained and controlled release application error is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of novel pesticide application, and in particular to a method and device for applying a slow-release pesticide. Background Art

[0002] Controlled-release pesticides are specially treated pesticides whose release rate and time can be precisely controlled so as to slowly release the active ingredients over a long period of time. This technology aims to improve the efficiency of pesticide use, reduce environmental pollution, and reduce potential harm to humans and other non-target organisms. Although the release rate of controlled-release pesticides can be controlled, unreasonable application areas, locations, and times of controlled-release pesticides will make it difficult to stably exert the efficacy of the drug, which will make it difficult to perform high-quality killing and control of pests during the planting process of crop plants, further excessively increase the scale of pest reproduction activities, and make crop plants more susceptible to pests. At the same time, unreasonable application of controlled-release pesticides will affect the healthy growth of crop plants, causing crop plants to wither and die, reducing their planting quality. Therefore, it is necessary to develop an application method that can improve the accuracy and stability of controlled-release pesticide application to solve the above problems. Summary of the invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a method and device for applying a slow-release pesticide.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] The first aspect of the present invention provides a method for applying a sustained-release pesticide, comprising the following steps:

[0006] S102: obtaining pest distribution data of the planting space to be applied with the slow-release pesticide in a predetermined continuous time series, obtaining a dynamic hotspot distribution map by calculating the kernel density estimation of the pest distribution data, planning the application area based on the dynamic hotspot distribution map, and generating an application area map;

[0007] S104: analyzing the degree of inhibition of the growth rate of plant varieties by the scale of the reproduction activities of the pest species through the application area map, and estimating and analyzing the application dosage based on the inhibition degree with the target planting requirements as the trend benchmark, to obtain the target application dosage of the slow-release pesticide required for each application area in the application area map;

[0008] S106: constructing a current trajectory decision tree, using the current trajectory decision tree to predict and fit the predicted breeding activity area of ​​the pest species in the multidimensional space, screening the application points of the breeding activity prediction area through the controlled release action range of the preset monomer dose of the controlled release pesticide, and obtaining the accurate application points of each controlled release pesticide;

[0009] S108: constructing a kurtosis trend model of hotspot distribution density-time frame, analyzing the time frame corresponding to the kurtosis point where the actual reproduction activity scale of the pest species exceeds the reasonable reproduction activity scale through the kurtosis trend model, and formulating a reasonable application time point of the slow-release pesticide based on the time frame number;

[0010] S110: Obtain multidimensional survival index parameters of the pest types by executing the target application dosage, accurate application points and reasonable application time points, determine the application error term that causes the abnormal multidimensional survival index to produce a survival index deviation based on the multidimensional survival index parameter analysis and optimize it, and obtain an application optimization plan for slow-release pesticides.

[0011] More specifically, the step S102 includes the following steps:

[0012] Acquire the planting space to be applied with the controlled-release pesticide and the prevention and control log of the planting space to be applied, and extract a number of pest distribution data of the planting space to be applied in a predetermined continuous time sequence through the prevention and control log;

[0013] Obtaining a preset monitoring interval for pests by a pest monitoring device in a planting space to be applied, setting a sliding time window according to the monitoring interval, and dividing the pest distribution data into sub-pest distribution data of multiple time segments through the sliding time window;

[0014] A kernel density estimation algorithm is introduced, a Gaussian kernel function and a bandwidth parameter of the Gaussian kernel function are preset in the kernel density estimation algorithm based on a given continuous time series, sub-pest distribution data of each time segment is calculated according to the Gaussian kernel function and the bandwidth parameter, and a kernel density value of each sub-pest distribution data is generated;

[0015] Obtain the kernel density value of all sub-pest distribution data, construct a heat map, estimate the density function of multi-time segmented pest distribution in the planting space to be applied based on the kernel density value of all sub-pest distribution data, and fit it in the heat map to generate a dynamic hot spot distribution map of the planting space to be applied;

[0016] A flash frequency threshold is preset, and hotspot distribution areas with flash frequencies greater than the flash frequency threshold are planned in the dynamic hotspot distribution map, defined as suspicious hotspot distribution areas, and target pest types for controlled-release pesticide control are obtained;

[0017] The benchmark pest hazard level of the target pest species is obtained based on the big data network, and the danger hotspot color threshold is preset according to the benchmark pest hazard level. If the danger hotspot color threshold can be queried in the suspicious hotspot distribution area, the suspicious hotspot distribution area is marked as the application area, and an application area map is generated.

[0018] More specifically, the step S104 includes the following steps:

[0019] Acquire image data of plants planted in each application area of ​​the application area map, introduce an edge detection algorithm to perform feature calculation and description on the image data, and obtain a plant phenotypic feature vector;

[0020] Retrieving plant variety information and current growth stage in a big data network based on the plant phenotypic feature vector, and extracting several historical growth rates of the plant at the current growth stage in the prevention and control log based on the plant variety information;

[0021] By analyzing the hotspot distribution changes in the application area in the dynamic hotspot distribution map, the reproduction activity scale of the target pest species on the plant variety at the current growth stage is obtained, and several hotspot density change values ​​of the target pest species under the reproduction activity scale are extracted;

[0022] A multivariate linear regression model is constructed based on the growth rate and the scale of reproductive activities, and the least square method is introduced to fit the multivariate linear regression model in the least square method through the several historical growth rates and several hot spot density change values ​​to output the regression coefficient;

[0023] Obtaining the preset monomer dosage of the slow-release pesticide and the target planting requirements, and determining the degree of inhibition of the growth rate of the plant variety by the scale of the reproduction activity of the target pest species according to the regression coefficient;

[0024] Taking the target planting requirement as the trend benchmark, the required preset monomer dosage is specifically estimated and analyzed based on the inhibition degree until it approaches the trend benchmark, and the estimation is terminated to finally lock in the target application dosage of the slow-release pesticide required for each application area in the application area map.

[0025] More specifically, the step S106 includes the following steps:

[0026] Construct a multidimensional space domain for each application area in the application area map, and obtain the spatial dynamic hot spot distribution of pest species in the multidimensional space domain on a continuous and uniform time frame through a dynamic hot spot distribution map;

[0027] According to the spatial dynamic hot spot distribution on a continuous uniform time frame, the trajectory changes of pest species in the multidimensional space are analyzed to determine multiple reproduction activity trajectories of pest species in the multidimensional space;

[0028] Preset a future time node, create a root node according to the future time node, use multiple reproductive activity trajectories to continuously recursively split nodes starting from the root node, and finally form a current trajectory decision tree;

[0029] Obtaining a current leaf node value of a current trajectory decision tree, calculating a time residual between the current leaf node value and the spatial dynamic hotspot distribution, retraining a new trajectory decision tree through the time residual, and obtaining a new leaf node value finally output by the new trajectory decision tree;

[0030] The new leaf node value is added to the current leaf node value to update the current trajectory decision tree, and the above steps are repeated to continuously iterate the current trajectory decision tree until the time residual is less than the preset time residual threshold, so as to obtain the predicted trajectory of the reproduction activity of the pest species when it reaches the future time node in the multidimensional space field;

[0031] Based on the spatial dynamic hotspot distribution, the breeding activity range of the pest species is preset, and the breeding activity range is fitted to the breeding activity prediction trajectory to obtain the breeding activity prediction area of ​​the pest species in the multidimensional space field;

[0032] Obtaining the preset monomer dosage of the controlled-release pesticide and the controlled-release efficacy range of the monomer dosage, calculating the monomer quantity of the controlled-release pesticide required for the application area in combination with the target application dosage and the preset monomer dosage, and defining the coverage threshold of the controlled-release efficacy range under the preset monomer quantity;

[0033] A backtracking algorithm is introduced to screen the application points when the sustained-release efficacy range of each preset monomer dose completely covers the predicted area of ​​reproductive activity based on the number of monomers until the coverage range definition threshold is met, thereby obtaining the accurate application points of each sustained-release pesticide in the application area.

[0034] More specifically, the step S108 includes the following steps:

[0035] Obtaining the hotspot distribution density values ​​corresponding to the pest species in the application area on multiple predetermined uniform time frames, constructing a kurtosis trend model, importing each of the hotspot distribution density values ​​into the kurtosis trend model for fitting and drawing, and obtaining a kurtosis trend model of hotspot distribution density-time frame;

[0036] Based on the big data network, the reasonable reproduction activity scale density coefficient of the pest species under different preset meteorological environmental parameters is obtained, and the actual meteorological environmental parameters of the current application area are obtained;

[0037] A hash algorithm is introduced to calculate a hash function between the actual meteorological environment parameter and each different preset meteorological environment parameter to obtain a plurality of hash functions, and a degree of agreement between the actual meteorological environment parameter and each different preset meteorological environment parameter is determined according to the plurality of hash functions;

[0038] Only the reasonable reproduction activity scale density coefficient corresponding to the preset meteorological environmental parameters with the maximum consistency is extracted, which is defined as the reasonable reproduction activity scale density coefficient of the pest species under the actual meteorological environmental conditions in the current application area;

[0039] The actual kurtosis trend model diagram of the pest species when it is in the application area is extracted through the kurtosis trend model of the hotspot distribution density-time frame, and a reasonable kurtosis trend model diagram is constructed according to the reasonable reproduction activity scale density coefficient of the pest species under the actual meteorological environment conditions in the current application area;

[0040] The kurtosis of the non-overlapping part between the actual kurtosis trend model diagram and the reasonable kurtosis trend model diagram is marked, which is defined as an abnormal kurtosis point. The time frame corresponding to the abnormal kurtosis point is queried through the kurtosis trend model, and the reasonable application time point of the sustained-release pesticide is determined according to the time frame.

[0041] More specifically, the step S110 includes the following steps:

[0042] The controlled-release pesticide is applied globally to the planting space to be applied by means of the target application dosage, accurate application point and reasonable application time point, and the multi-dimensional survival index parameters of the pest types are recorded;

[0043] Obtaining preset control requirements for pest types, and obtaining the minimum standard critical threshold of each multidimensional survival indicator through the preset control requirements;

[0044] Constructing a survival indicator radar chart, setting and planning a minimum standard radar area of ​​the multidimensional survival indicator in the survival indicator radar chart according to the minimum standard critical threshold of each multidimensional survival indicator, and importing the multidimensional survival indicator parameters into the survival indicator radar chart for drawing, so as to obtain an actual radar area of ​​the multidimensional survival indicator;

[0045] Eliminating the overlapping radar areas of various multidimensional survival indicators between the minimum standard radar area and the actual radar area, obtaining non-overlapping radar areas of various multidimensional survival indicators, and obtaining area values ​​of the non-overlapping radar areas;

[0046] If the area value of the region is greater than the preset area value of the region, the multidimensional survival index corresponding to the area value of the region is calibrated as an abnormal multidimensional survival index, and the target administration dose, accurate administration point and reasonable administration time point are defined as administration error terms;

[0047] Obtaining the actual index value of the abnormal multidimensional survival index, calculating the difference between the actual index value and the minimum standard critical threshold value corresponding to the abnormal multidimensional survival index, obtaining the index deviation value, introducing the fuzzy C-means clustering algorithm to cluster the index deviation values ​​existing in each abnormal multidimensional survival index into each application error term, and obtaining the application error fuzzy clustering label of each abnormal multidimensional survival index;

[0048] The application error term corresponding to the survival indicator deviation of each abnormal multidimensional survival indicator is determined according to the application error fuzzy clustering label, and the application error term is subjected to application adjustment optimization to obtain an application optimization plan for the slow-release pesticide.

[0049] A second aspect of the present invention provides an application device for a slow-release pesticide, the application device comprising a memory and a processor, the memory storing a program for an application method for a slow-release pesticide, and when the program for an application method for a slow-release pesticide is executed by the processor, any one of the steps for an application method for a slow-release pesticide is implemented.

[0050] The present invention solves the technical defects existing in the background technology, and the beneficial technical effects of the present invention are:

[0051] A dynamic hotspot distribution map is constructed by acquiring the pest distribution data of the planting space to be applied with slow-release pesticides in a predetermined continuous time series, and the application area is planned based on the dynamic hotspot distribution map to generate an application area map; the degree of inhibition of the scale of the reproduction activities of pest species on the growth rate of plant varieties is analyzed through the application area map, and the application dosage is estimated and analyzed based on the inhibition degree to obtain the target application dosage of the slow-release pesticide required for each application area in the application area map; a current trajectory decision tree is constructed, and the current trajectory decision tree is used to predict and fit the predicted area of ​​the reproduction activities of pest species in the multidimensional space field, and the slow-release pesticide preset monomer dosage is used to predict the slow-release pesticide. The scope of action screens the application points in the predicted area of ​​reproduction activities to obtain the accurate application points of each controlled-release pesticide; constructs a kurtosis trend model of hotspot distribution density-time frame, and uses the kurtosis trend model to analyze the time frame corresponding to the kurtosis point where the actual reproduction activity scale of the pest species exceeds the reasonable reproduction activity scale, and formulates a reasonable application time point for the controlled-release pesticide based on the number of time frames; by executing the target application dose, accurate application point, and multidimensional survival index parameter analysis after the reasonable application time point, the application error term that causes the abnormal multidimensional survival index to produce a survival index deviation is determined and optimized to obtain an application optimization scheme for the controlled-release pesticide. The present invention can accurately and reasonably plan and optimize the application of controlled-release pesticides to improve the quality of the effect of controlled-release pesticides on pests and reduce the application error of controlled-release pesticides. 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 or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0053] Figure 1 A first method flow chart of a method for applying a slow-release pesticide is shown;

[0054] Figure 2 A second method flow chart of a method for applying a slow-release pesticide is shown;

[0055] Figure 3 The device structure diagram of a device for applying a slow-release pesticide is shown. DETAILED DESCRIPTION

[0056] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0057] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0058] The first aspect of the present invention provides a method for applying a sustained-release pesticide, such as Figure 1 As shown, the following steps are included:

[0059] S102: obtaining pest distribution data of the planting space to be applied with the slow-release pesticide in a predetermined continuous time series, obtaining a dynamic hotspot distribution map by calculating the kernel density estimation of the pest distribution data, planning the application area based on the dynamic hotspot distribution map, and generating an application area map;

[0060] S104: analyzing the degree of inhibition of the growth rate of plant varieties by the scale of the reproduction activities of the pest species through the application area map, and estimating and analyzing the application dosage based on the inhibition degree with the target planting requirements as the trend benchmark, to obtain the target application dosage of the slow-release pesticide required for each application area in the application area map;

[0061] S106: constructing a current trajectory decision tree, using the current trajectory decision tree to predict and fit the predicted breeding activity area of ​​the pest species in the multidimensional space, screening the application points of the breeding activity prediction area through the controlled release action range of the preset monomer dose of the controlled release pesticide, and obtaining the accurate application points of each controlled release pesticide;

[0062] S108: constructing a kurtosis trend model of hotspot distribution density-time frame, analyzing the time frame corresponding to the kurtosis point where the actual reproduction activity scale of the pest species exceeds the reasonable reproduction activity scale through the kurtosis trend model, and formulating a reasonable application time point of the slow-release pesticide based on the time frame number;

[0063] S110: Obtain multidimensional survival index parameters of the pest types by executing the target application dosage, accurate application points and reasonable application time points, determine the application error term that causes the abnormal multidimensional survival index to produce a survival index deviation based on the multidimensional survival index parameter analysis and optimize it, and obtain an application optimization plan for slow-release pesticides.

[0064] More specifically, the step S102 is as follows: Figure 2 As shown, the specific steps include:

[0065] S202: Acquire the planting space to be applied with the slow-release pesticide and the control log of the planting space to be applied, and extract a plurality of insect pest distribution data of the planting space to be applied in a predetermined continuous time sequence through the control log;

[0066] S204: obtaining a preset monitoring interval for pests by the pest monitoring device in the planting space to be applied, setting a sliding time window according to the monitoring interval, and dividing the pest distribution data into sub-pest distribution data of multiple time segments through the sliding time window;

[0067] S206: introducing a kernel density estimation algorithm, presetting a Gaussian kernel function and a bandwidth parameter of the Gaussian kernel function in the kernel density estimation algorithm based on a predetermined continuous time series, calculating the sub-pest distribution data of each time segment according to the Gaussian kernel function and the bandwidth parameter, and generating a kernel density value of each sub-pest distribution data;

[0068] S208: Obtain the kernel density values ​​of all sub-pest distribution data, construct a heat map, estimate the density function of the multi-time segmented pest distribution in the planting space to be applied based on the kernel density values ​​of all sub-pest distribution data, and fit it in the heat map to generate a dynamic hot spot distribution map of the planting space to be applied;

[0069] S210: Preset a flash frequency threshold, plan a hotspot distribution area with a flash frequency greater than the flash frequency threshold in the dynamic hotspot distribution map, define it as a suspicious hotspot distribution area, and obtain the target pest type for controlled-release pesticide control;

[0070] S212: Obtain a baseline pest hazard level for the target pest species based on a big data network, preset a danger hotspot color threshold based on the baseline pest hazard level, and if the danger hotspot color threshold can be queried in the suspicious hotspot distribution area, mark the suspicious hotspot distribution area as an application area, and generate an application area map.

[0071] It should be noted that controlled-release pesticides can effectively and continuously control pests and pathogens. In order to make controlled-release pesticides play a higher control value, controlled-release pesticides need to be applied to areas with more pests in the planting space; however, the traditional controlled-release pesticide application method is usually manual random application, which makes the controlled-release pesticide distribution extremely uneven, so that areas with more pests may not be applied, while areas with less pests have a concentration of controlled-release pesticides that exceeds the standard, making it difficult for controlled-release pesticides to play the expected killing and control effect, resulting in a waste of pesticides. Therefore, through this method, a reasonable application area can be selected for the application of controlled-release pesticides according to the distribution of pest hotspots in the planting area to be applied, thereby further improving the control value and pest control performance of controlled-release pesticides, avoiding poor pest control effects and wasteful cost output caused by uneven application, and greatly increasing economic benefits.

[0072] More specifically, the step S104 includes the following steps:

[0073] Acquire image data of plants planted in each application area of ​​the application area map, introduce an edge detection algorithm to perform feature calculation and description on the image data, and obtain a plant phenotypic feature vector;

[0074] Retrieving plant variety information and current growth stage in a big data network based on the plant phenotypic feature vector, and extracting several historical growth rates of the plant at the current growth stage in the prevention and control log based on the plant variety information;

[0075] By analyzing the hotspot distribution changes in the application area in the dynamic hotspot distribution map, the reproduction activity scale of the target pest species on the plant variety at the current growth stage is obtained, and several hotspot density change values ​​of the target pest species under the reproduction activity scale are extracted;

[0076] A multivariate linear regression model is constructed based on the growth rate and the scale of reproductive activities, and the least square method is introduced to fit the multivariate linear regression model in the least square method through the several historical growth rates and several hot spot density change values ​​to output the regression coefficient;

[0077] Obtaining the preset monomer dosage of the slow-release pesticide and the target planting requirements, and determining the degree of inhibition of the growth rate of the plant variety by the scale of the reproduction activity of the target pest species according to the regression coefficient;

[0078] Taking the target planting requirement as the trend benchmark, the required preset monomer dosage is specifically estimated and analyzed based on the inhibition degree until it approaches the trend benchmark, and the estimation is terminated to finally lock in the target application dosage of the slow-release pesticide required for each application area in the application area map.

[0079] It should be noted that the application dosage of slow-release pesticides is crucial. Although a higher application dosage can improve the pest control effect, it will also harm crop plants, causing crop plants to wither and necrotize. If the application dosage is lower, the pest control performance may be poor, causing excessive reproduction and activity migration of pests, thereby causing crop plants to be eroded by a large number of pests, reducing the planting quality and yield of crop plants. This method can calculate the degree of inhibition of the growth rate of plant varieties by the scale of pest species reproduction activities, and then accurately estimate and measure the optimal application dosage of slow-release pesticides based on the inhibition degree, thereby ensuring the accuracy of the application of slow-release pesticides and the full play of their efficacy, improving the rationality and safety factor of the application of slow-release pesticides, effectively protecting crop plants and maximally controlling the reproduction and activity of pests, and optimizing the release rate of slow-release pesticides.

[0080] More specifically, the step S106 includes the following steps:

[0081] Construct a multidimensional space domain for each application area in the application area map, and obtain the spatial dynamic hot spot distribution of pest species in the multidimensional space domain on a continuous and uniform time frame through a dynamic hot spot distribution map;

[0082] According to the spatial dynamic hot spot distribution on a continuous uniform time frame, the trajectory changes of pest species in the multidimensional space are analyzed to determine multiple reproduction activity trajectories of pest species in the multidimensional space;

[0083] Preset a future time node, create a root node according to the future time node, use multiple reproductive activity trajectories to continuously recursively split nodes starting from the root node, and finally form a current trajectory decision tree;

[0084] Obtaining a current leaf node value of a current trajectory decision tree, calculating a time residual between the current leaf node value and the spatial dynamic hotspot distribution, retraining a new trajectory decision tree through the time residual, and obtaining a new leaf node value finally output by the new trajectory decision tree;

[0085] The new leaf node value is added to the current leaf node value to update the current trajectory decision tree, and the above steps are repeated to continuously iterate the current trajectory decision tree until the time residual is less than the preset time residual threshold, so as to obtain the predicted trajectory of the reproduction activity of the pest species when it reaches the future time node in the multidimensional space field;

[0086] Based on the spatial dynamic hotspot distribution, the breeding activity range of the pest species is preset, and the breeding activity range is fitted to the breeding activity prediction trajectory to obtain the breeding activity prediction area of ​​the pest species in the multidimensional space field;

[0087] Obtaining the preset monomer dosage of the controlled-release pesticide and the controlled-release efficacy range of the monomer dosage, calculating the monomer quantity of the controlled-release pesticide required for the application area in combination with the target application dosage and the preset monomer dosage, and defining the coverage threshold of the controlled-release efficacy range under the preset monomer quantity;

[0088] A backtracking algorithm is introduced to screen the application points when the sustained-release efficacy range of each preset monomer dose completely covers the predicted area of ​​reproductive activity based on the number of monomers until the coverage range definition threshold is met, thereby obtaining the accurate application points of each sustained-release pesticide in the application area.

[0089] It should be noted that after the specific application dosage of the slow-release pesticide is determined, it can be applied in the application area. However, the distribution and reproduction activities of crop plants and pests in different application areas are not uniform, and there are large differences between different application areas. If the slow-release pesticide is applied uniformly to the same fixed point, it may be difficult for the efficacy range of the slow-release pesticide to fully cover the static distribution range and dynamic reproduction activity range of pests in the application area, resulting in efficacy deviations of the unreasonable application of slow-release pesticides, making it difficult to achieve accurate and efficient killing and control effects. Therefore, this method can first predict the predicted breeding activity area of ​​pests in the future, and then screen the optimal application points based on the predicted breeding activity area in the future combined with the preset monomer dose of the slow-release pesticide itself and the slow-release efficacy range to cover the application area. This can further improve the global coverage rate of the efficacy of the slow-release pesticide on pests, greatly improve the pest control effect, avoid the efficacy deviation of the slow-release pesticide pest control, enable the slow-release pesticide to fully exert its efficacy, improve the stability of the slow-release rate, and have high reliability.

[0090] More specifically, the step S108 includes the following steps:

[0091] Obtaining the hotspot distribution density values ​​corresponding to the pest species in the application area on multiple predetermined uniform time frames, constructing a kurtosis trend model, importing each of the hotspot distribution density values ​​into the kurtosis trend model for fitting and drawing, and obtaining a kurtosis trend model of hotspot distribution density-time frame;

[0092] Based on the big data network, the reasonable reproduction activity scale density coefficient of the pest species under different preset meteorological environmental parameters is obtained, and the actual meteorological environmental parameters of the current application area are obtained;

[0093] A hash algorithm is introduced to calculate a hash function between the actual meteorological environment parameter and each different preset meteorological environment parameter to obtain a plurality of hash functions, and a degree of agreement between the actual meteorological environment parameter and each different preset meteorological environment parameter is determined according to the plurality of hash functions;

[0094] Only the reasonable reproduction activity scale density coefficient corresponding to the preset meteorological environmental parameters with the maximum consistency is extracted, which is defined as the reasonable reproduction activity scale density coefficient of the pest species under the actual meteorological environmental conditions in the current application area;

[0095] The actual kurtosis trend model diagram of the pest species when it is in the application area is extracted through the kurtosis trend model of the hotspot distribution density-time frame, and a reasonable kurtosis trend model diagram is constructed according to the reasonable reproduction activity scale density coefficient of the pest species under the actual meteorological environment conditions in the current application area;

[0096] The kurtosis of the non-overlapping part between the actual kurtosis trend model diagram and the reasonable kurtosis trend model diagram is marked, which is defined as an abnormal kurtosis point. The time frame corresponding to the abnormal kurtosis point is queried through the kurtosis trend model, and the reasonable application time point of the sustained-release pesticide is determined according to the time frame.

[0097] It should be noted that the meteorological environment includes sunny days, cloudy days, rainy days, temperature, humidity, light and oxygen content. When an unreasonable scale of reproduction activity of pests appears in crop plants, it is necessary to kill and control them by applying slow-release pesticides. The unreasonable scale of reproduction activity of pests changes with the change of meteorological environmental conditions, and meteorological environmental conditions change with time. Therefore, this method can analyze whether the pests in the current application area have an unreasonable scale of reproduction activity through the kurtosis trend of pest distribution. If it appears, slow-release pesticides can be applied at this time point, realizing the rationalization of slow-release pesticides. Compared with the traditional application time point selection method, this method determines the application time according to the change of the scale of pest reproduction activity under meteorological environmental factors, greatly improving the application accuracy of slow-release pesticides, preventing the failure to apply pesticides when the scale of pest reproduction activity is large and the application of a large amount of slow-release pesticides when the scale of reproduction activity is small, ensuring high-quality killing and control of pests, and protecting crop plants to the maximum extent.

[0098] More specifically, the step S110 includes the following steps:

[0099] The controlled-release pesticide is applied globally to the planting space to be applied by means of the target application dosage, accurate application point and reasonable application time point, and the multi-dimensional survival index parameters of the pest types are recorded;

[0100] Obtaining preset control requirements for pest types, and obtaining the minimum standard critical threshold of each multidimensional survival indicator through the preset control requirements;

[0101] Constructing a survival indicator radar chart, setting and planning a minimum standard radar area of ​​the multidimensional survival indicator in the survival indicator radar chart according to the minimum standard critical threshold of each multidimensional survival indicator, and importing the multidimensional survival indicator parameters into the survival indicator radar chart for drawing, so as to obtain an actual radar area of ​​the multidimensional survival indicator;

[0102] Eliminating the overlapping radar areas of various multidimensional survival indicators between the minimum standard radar area and the actual radar area, obtaining non-overlapping radar areas of various multidimensional survival indicators, and obtaining area values ​​of the non-overlapping radar areas;

[0103] If the area value of the region is greater than the preset area value of the region, the multidimensional survival index corresponding to the area value of the region is calibrated as an abnormal multidimensional survival index, and the target administration dose, accurate administration point and reasonable administration time point are defined as administration error terms;

[0104] Obtaining the actual index value of the abnormal multidimensional survival index, calculating the difference between the actual index value and the minimum standard critical threshold value corresponding to the abnormal multidimensional survival index, obtaining the index deviation value, introducing the fuzzy C-means clustering algorithm to cluster the index deviation values ​​existing in each abnormal multidimensional survival index into each application error term, and obtaining the application error fuzzy clustering label of each abnormal multidimensional survival index;

[0105] The application error term corresponding to the survival indicator deviation of each abnormal multidimensional survival indicator is determined according to the application error fuzzy clustering label, and the application error term is subjected to application adjustment optimization to obtain an application optimization plan for the slow-release pesticide.

[0106] It should be noted that the multidimensional survival index parameters include population survival rate parameters, juvenile survival rate parameters, disease-free survival rate parameters, average survival time and mortality rate. There may be certain errors in the application of controlled-release pesticides based on the target application dose, accurate application points and reasonable application time points. Considering that these errors will reduce the control effect of controlled-release pesticides and cause a lot of waste, this method can accurately analyze the application effect of controlled-release pesticides based on the survival of pests in the application area after application based on the target application dose, accurate application points and reasonable application time points, and determine the application error items corresponding to the deviation of the survival index for each abnormal multidimensional survival index according to the abnormal multidimensional survival index, so as to achieve targeted adjustment and optimization, improve the stability and reliability of the application quality of controlled-release pesticides, and eliminate application errors to the greatest extent.

[0107] In addition, the method for applying a sustained-release pesticide further comprises the following steps:

[0108] Based on the big data network, the application methods associated with the oil spill of the submarine oil-filled cable are obtained, and the historical application loss rate corresponding to each application method is obtained;

[0109] If the historical application loss rate is less than the preset application loss rate, then the application means corresponding to the historical application loss rate being less than the preset application loss rate are extracted to obtain the application means after the initial screening;

[0110] Obtaining the pest control scale of the current planting area to be applied, introducing the Spearson correlation coefficient to calculate the correlation coefficient between the pest control scale and each of the application means after the initial screening, and determining the application suitability of each application means to the pest control scale according to the correlation coefficient;

[0111] Eliminate the application means after the primary screening whose application suitability does not exceed the preset application suitability, and integrate the remaining application means to obtain the application means after the secondary screening;

[0112] Obtaining the historical application loss rates corresponding to the application means after the secondary screening, constructing a descending sorting table, and importing the historical application loss rates corresponding to the application means after the secondary screening into the descending sorting table for descending sorting;

[0113] After the descending order is completed, the secondary screened application method corresponding to the minimum historical application loss rate is extracted as the designated application plan for the current planting area to be applied.

[0114] It should be noted that there are many ways to apply slow-release pesticides, for example, the pesticides are wrapped in biodegradable films or polymers to form slow-release particles for application, or the slow-release pesticides are combined with drip irrigation systems, and the pesticides are evenly released into the soil through the drip irrigation system. This method can select a targeted and low-loss application method for the current pest control scale of the planting area to be applied for prevention and control. On the one hand, it can not only achieve the best prevention and control effect on the basis of accurately controlling the release amount of the pesticide, but also reduce the waste of the pesticide, improve the healthy growth of crop plants, and have high economic benefits and reliability.

[0115] The second aspect of the present invention provides a device for applying a controlled-release pesticide, such as Figure 3 As shown, the application device of a controlled-release pesticide includes a memory 31 and a processor 32. The memory 31 stores a program for the application method of a controlled-release pesticide. When the program for the application method of a controlled-release pesticide is executed by the processor 32, any one of the steps for the application method of a controlled-release pesticide is implemented.

[0116] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for applying a sustained-release pesticide, characterized in that: The following steps are involved: S102: obtaining pest distribution data of the planting space to be applied with the slow-release pesticide in a predetermined continuous time series, obtaining a dynamic hotspot distribution map by calculating the kernel density estimation of the pest distribution data, planning the application area based on the dynamic hotspot distribution map, and generating an application area map; S104: analyzing the degree of inhibition of the growth rate of plant varieties by the scale of the reproduction activities of the pest species through the application area map, and estimating and analyzing the application dosage based on the inhibition degree with the target planting requirements as the trend benchmark, to obtain the target application dosage of the slow-release pesticide required for each application area in the application area map; S106: constructing a current trajectory decision tree, using the current trajectory decision tree to predict and fit the predicted breeding activity area of ​​the pest species in the multidimensional space, screening the application points of the breeding activity prediction area through the controlled release action range of the preset monomer dose of the controlled release pesticide, and obtaining the accurate application points of each controlled release pesticide; S108: constructing a kurtosis trend model of hotspot distribution density-time frame, analyzing the time frame corresponding to the kurtosis point where the actual reproduction activity scale of the pest species exceeds the reasonable reproduction activity scale through the kurtosis trend model, and formulating a reasonable application time point of the slow-release pesticide based on the time frame number; S110: Obtain multidimensional survival index parameters of the pest types by executing the target application dosage, accurate application points and reasonable application time points, determine the application error term that causes the abnormal multidimensional survival index to produce a survival index deviation based on the multidimensional survival index parameter analysis and optimize it, and obtain an application optimization plan for slow-release pesticides.

2. The method for applying a sustained-release pesticide according to claim 1, characterized in that: The step S102 specifically includes the following steps: Acquire the planting space to be applied with the controlled-release pesticide and the prevention and control log of the planting space to be applied, and extract a number of pest distribution data of the planting space to be applied in a predetermined continuous time sequence through the prevention and control log; Obtaining a preset monitoring interval for pests by a pest monitoring device in a planting space to be applied, setting a sliding time window according to the monitoring interval, and dividing the pest distribution data into sub-pest distribution data of multiple time segments through the sliding time window; A kernel density estimation algorithm is introduced, a Gaussian kernel function and a bandwidth parameter of the Gaussian kernel function are preset in the kernel density estimation algorithm based on a given continuous time series, sub-pest distribution data of each time segment is calculated according to the Gaussian kernel function and the bandwidth parameter, and a kernel density value of each sub-pest distribution data is generated; Obtain the kernel density value of all sub-pest distribution data, construct a heat map, estimate the density function of multi-time segmented pest distribution in the planting space to be applied based on the kernel density value of all sub-pest distribution data, and fit it in the heat map to generate a dynamic hot spot distribution map of the planting space to be applied; A flash frequency threshold is preset, and hotspot distribution areas with flash frequencies greater than the flash frequency threshold are planned in the dynamic hotspot distribution map, defined as suspicious hotspot distribution areas, and target pest types for controlled-release pesticide control are obtained; The benchmark pest hazard level of the target pest species is obtained based on the big data network, and the danger hotspot color threshold is preset according to the benchmark pest hazard level. If the danger hotspot color threshold can be queried in the suspicious hotspot distribution area, the suspicious hotspot distribution area is marked as the application area, and an application area map is generated.

3. The method for applying a sustained-release pesticide according to claim 1, characterized in that: The step S104 specifically includes the following steps: Acquire image data of plants planted in each application area of ​​the application area map, introduce an edge detection algorithm to perform feature calculation and description on the image data, and obtain a plant phenotypic feature vector; Retrieving plant variety information and current growth stage in a big data network based on the plant phenotypic feature vector, and extracting several historical growth rates of the plant at the current growth stage in the prevention and control log based on the plant variety information; By analyzing the hotspot distribution changes in the application area in the dynamic hotspot distribution map, the reproduction activity scale of the target pest species on the plant variety at the current growth stage is obtained, and several hotspot density change values ​​of the target pest species under the reproduction activity scale are extracted; A multivariate linear regression model is constructed based on the growth rate and the scale of reproductive activities, and the least square method is introduced to fit the multivariate linear regression model in the least square method through the several historical growth rates and several hot spot density change values ​​to output the regression coefficient; Obtaining the preset monomer dosage of the slow-release pesticide and the target planting requirements, and determining the degree of inhibition of the growth rate of the plant variety by the scale of the reproduction activity of the target pest species according to the regression coefficient; Taking the target planting requirement as the trend benchmark, the required preset monomer dosage is specifically estimated and analyzed based on the inhibition degree until it approaches the trend benchmark, and the estimation is terminated to finally lock in the target application dosage of the slow-release pesticide required for each application area in the application area map.

4. The method for applying a sustained-release pesticide according to claim 1, characterized in that: The step S106 specifically includes the following steps: Construct a multidimensional space domain for each application area in the application area map, and obtain the spatial dynamic hot spot distribution of pest species in the multidimensional space domain on a continuous and uniform time frame through a dynamic hot spot distribution map; According to the spatial dynamic hot spot distribution on a continuous uniform time frame, the trajectory changes of pest species in the multidimensional space are analyzed to determine multiple reproduction activity trajectories of pest species in the multidimensional space; Preset a future time node, create a root node according to the future time node, use multiple reproductive activity trajectories to continuously recursively split nodes starting from the root node, and finally form a current trajectory decision tree; Obtaining a current leaf node value of a current trajectory decision tree, calculating a time residual between the current leaf node value and the spatial dynamic hotspot distribution, retraining a new trajectory decision tree through the time residual, and obtaining a new leaf node value finally output by the new trajectory decision tree; The new leaf node value is added to the current leaf node value to update the current trajectory decision tree, and the above steps are repeated to continuously iterate the current trajectory decision tree until the time residual is less than the preset time residual threshold, so as to obtain the predicted trajectory of the reproduction activity of the pest species when it reaches the future time node in the multidimensional space field; Based on the spatial dynamic hotspot distribution, the breeding activity range of the pest species is preset, and the breeding activity range is fitted to the breeding activity prediction trajectory to obtain the breeding activity prediction area of ​​the pest species in the multidimensional space field; Obtaining the preset monomer dosage of the controlled-release pesticide and the controlled-release efficacy range of the monomer dosage, calculating the monomer quantity of the controlled-release pesticide required for the application area in combination with the target application dosage and the preset monomer dosage, and defining the coverage threshold of the controlled-release efficacy range under the preset monomer quantity; A backtracking algorithm is introduced to screen the application points when the sustained-release efficacy range of each preset monomer dose completely covers the predicted area of ​​reproductive activity based on the number of monomers until the coverage range definition threshold is met, thereby obtaining the accurate application points of each sustained-release pesticide in the application area.

5. The method for applying a sustained-release pesticide according to claim 1, characterized in that: The step S108 specifically includes the following steps: Obtaining the hotspot distribution density values ​​corresponding to the pest species in the application area on multiple predetermined uniform time frames, constructing a kurtosis trend model, importing each of the hotspot distribution density values ​​into the kurtosis trend model for fitting and drawing, and obtaining a kurtosis trend model of hotspot distribution density-time frame; Based on the big data network, the reasonable reproduction activity scale density coefficient of the pest species under different preset meteorological environmental parameters is obtained, and the actual meteorological environmental parameters of the current application area are obtained; A hash algorithm is introduced to calculate a hash function between the actual meteorological environment parameter and each different preset meteorological environment parameter to obtain a plurality of hash functions, and a degree of agreement between the actual meteorological environment parameter and each different preset meteorological environment parameter is determined according to the plurality of hash functions; Only the reasonable reproduction activity scale density coefficient corresponding to the preset meteorological environmental parameters with the maximum consistency is extracted, which is defined as the reasonable reproduction activity scale density coefficient of the pest species under the actual meteorological environmental conditions in the current application area; The actual kurtosis trend model diagram of the pest species when it is in the application area is extracted through the kurtosis trend model of the hotspot distribution density-time frame, and a reasonable kurtosis trend model diagram is constructed according to the reasonable reproduction activity scale density coefficient of the pest species under the actual meteorological environment conditions in the current application area; The kurtosis of the non-overlapping part between the actual kurtosis trend model diagram and the reasonable kurtosis trend model diagram is marked, which is defined as an abnormal kurtosis point. The time frame corresponding to the abnormal kurtosis point is queried through the kurtosis trend model, and the reasonable application time point of the sustained-release pesticide is determined according to the time frame.

6. The method for applying a sustained-release pesticide according to claim 1, characterized in that: The step S110 specifically includes the following steps: The controlled-release pesticide is applied globally to the planting space to be applied by means of the target application dosage, accurate application point and reasonable application time point, and the multi-dimensional survival index parameters of the pest types are recorded; Obtaining preset control requirements for pest types, and obtaining the minimum standard critical threshold of each multidimensional survival indicator through the preset control requirements; Constructing a survival indicator radar chart, setting and planning a minimum standard radar area of ​​the multidimensional survival indicator in the survival indicator radar chart according to the minimum standard critical threshold of each multidimensional survival indicator, and importing the multidimensional survival indicator parameters into the survival indicator radar chart for drawing, so as to obtain an actual radar area of ​​the multidimensional survival indicator; Eliminating the overlapping radar areas of various multidimensional survival indicators between the minimum standard radar area and the actual radar area, obtaining non-overlapping radar areas of various multidimensional survival indicators, and obtaining area values ​​of the non-overlapping radar areas; If the area value of the region is greater than the preset area value of the region, the multidimensional survival index corresponding to the area value of the region is calibrated as an abnormal multidimensional survival index, and the target administration dose, accurate administration point and reasonable administration time point are defined as administration error terms; Obtaining the actual index value of the abnormal multidimensional survival index, calculating the difference between the actual index value and the minimum standard critical threshold value corresponding to the abnormal multidimensional survival index, obtaining the index deviation value, introducing the fuzzy C-means clustering algorithm to cluster the index deviation values ​​existing in each abnormal multidimensional survival index into each application error term, and obtaining the application error fuzzy clustering label of each abnormal multidimensional survival index; The application error term corresponding to the survival indicator deviation of each abnormal multidimensional survival indicator is determined according to the application error fuzzy clustering label, and the application error term is subjected to application adjustment optimization to obtain an application optimization plan for the slow-release pesticide.

7. A device for applying a controlled-release pesticide, characterized in that: The device for applying a controlled-release pesticide comprises a memory and a processor, wherein the memory stores a program for applying a controlled-release pesticide. When the program for applying a controlled-release pesticide is executed by the processor, the steps for applying a controlled-release pesticide as described in any one of claims 1 to 6 are implemented.

Images