A tobacco pest monitoring and early warning method and system based on the Internet of Things

Abstract

The application discloses a tobacco pest monitoring and early warning method and system based on the Internet of Things, relates to the technical field of pest monitoring, and comprises the following steps: taking a non-control loss curve as a benchmark, reading control measures and control costs from a tobacco field management resource account book, simulating the control measures at a candidate starting moment, obtaining post-control loss and control net income, screening candidate starting moments that meet the conditions of the control net income being not less than zero and the sum of the post-control loss and the control cost of the corresponding control measures being not more than the residual loss budget from the candidate starting moments according to the residual loss budget, the post-control loss and the control net income, defining an economically executable window based on the screened candidate starting moments, screening a resource reachable moment and calculating disposal margin after matching the control resources and the economically executable window, and obtaining a tobacco pest early warning and control priority order; and the scientificity, accuracy and operability of tobacco pest prevention and control are improved by calculating the disposal margin.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pest monitoring, in particular to a tobacco pest monitoring and early warning method and system based on the Internet of Things. BACKGROUND

[0002] As an important part of new generation information technology, the Internet of Things technology has received extensive attention and rapid development in the field of agricultural pest monitoring in recent years. With the continuous maturity of sensor technology, wireless communication technology and edge computing technology, the pest monitoring terminal based on the Internet of Things has been able to realize the wide range, high frequency and all-weather real-time collection and transmission of multi-source heterogeneous data such as temperature and humidity, light intensity, soil parameters and pest image, breaking through the limitations of the previous manual inspection method in terms of coverage, collection frequency and real-time performance. At the same time, the introduction of time series analysis, machine learning, geographic information (GIS) spatial interpolation and other data-driven methods enables the massive monitoring data collected by the Internet of Things to effectively support pest occurrence trend prediction and loss assessment, promoting a profound change from experience-driven to data-driven in the paradigm of tobacco pest monitoring and laying a solid information foundation and technical condition for building an intelligent and precise pest early warning system.

[0003] However, the existing tobacco pest monitoring and early warning method still has room for improvement. First, it fails to fully consider the migration and diffusion behavior of pests among different tobacco field plots, especially ignoring the influence of environmental factors such as wind direction and wind speed on pest migration path and migration weight, resulting in insufficient accuracy of pest transmission situation prediction. Second, it lacks an economic threshold decision mechanism to quantitatively correlate pest pressure and economic loss, making it impossible to provide optimal prevention and control timing recommendations based on cost and benefit analysis, and failing to incorporate prevention and control resource availability constraints into the early warning decision-making process, making it difficult to balance prevention and control effectiveness and economic benefits. SUMMARY

[0004] In view of the above existing problems, the present application is proposed.

[0005] Therefore, the present application provides a tobacco pest monitoring and early warning method based on the Internet of Things to solve the problem of insufficient accuracy of pest transmission situation prediction.

[0006] To solve the above technical problems, the present application provides the following technical solutions: In a first aspect, the present application provides a tobacco pest monitoring and early warning method based on the Internet of Things, which comprises, After collecting the boundaries of tobacco planting areas, pest management parameters, tobacco field management data, prevention and control measures and prevention and control resources, the tobacco field plots are divided and numbered, and a tobacco field management resource account book is established; Collect pest data, tobacco damage data, environmental data and terminal state data according to the tobacco field number and align them to generate a tobacco pest event sequence; The time sequence response check is performed on the tobacco field pest event sequence to obtain a credible pest pressure; The adjacent relationship of the tobacco field is identified, and the migration weight is calculated by combining the pest management parameters and the pest event sequence of the tobacco field, and the credible pest pressure is recursively propagated to obtain a propagation corrected pest pressure sequence; The residual loss budget is calculated according to the tobacco field management resource account, and the propagation corrected pest pressure sequence is converted into a no-control loss curve according to the pest management parameters; Taking the no-control loss curve as a reference, the control measures and control costs are read from the tobacco field management resource account, and the control measures are simulated at the candidate starting time to obtain the post-control loss and control net income; According to the residual loss budget, the post-control loss and the control net income, the candidate starting time that meets the condition that the control net income is not less than zero and the sum of the post-control loss and the control cost of the corresponding control measure is not more than the residual loss budget is screened from the candidate starting time, and the economic executable window is defined based on the screened candidate starting time; After matching the control resources and the economic executable window, the resource reachable time is screened and the disposal margin is calculated to obtain the tobacco pest warning and control priority.

[0007] As a preferred scheme of the tobacco pest monitoring and warning method based on the Internet of Things, the pest management parameters include the effective migration distance of the target pest, the migration attenuation coefficient and the pest pressure loss conversion coefficient; The tobacco field management data includes the expected yield value of the tobacco field, the minimum income preservation value, the occurred pest loss, the occurred control cost, the control cost corresponding to each control measure and the expected harvesting time; The control resources include the material demand and the available amount of material, the personnel working hours demand and the available working hours of personnel, the equipment use time and the available time period of equipment corresponding to each control measure.

[0008] As a preferred scheme of the tobacco pest monitoring and warning method based on the Internet of Things, the generation of the tobacco field pest event sequence is specifically: Taking the tobacco field number as a spatial index, the pest data, the tobacco damage data, the environment data and the terminal state data are synchronously collected in each tobacco field corresponding to the tobacco field number, wherein the pest data records the number of target pests and the collection time, the tobacco damage data records the damaged area and the collection time, the environment data records the wind speed, the wind direction and the collection time, and the terminal state data records the power supply state, the communication state, the collection state and the collection time; The pest data, the tobacco leaf damage data, the environment data and the terminal state data are collected according to the tobacco field numbers, the data of different collection time points are sorted with the collection time points as the time reference, the pest data, the tobacco leaf damage data, the environment data and the terminal state data of the same collection time point are matched to the same time position, the data of different collection time points are calculated for the collection time point difference, the data with the minimum collection time point difference is matched to the same time position, and then the data is continuously arranged from early to late according to the collection time point, so that the tobacco field pest situation event sequence corresponding to each tobacco field number is generated.

[0009] As a preferred scheme of the tobacco pest monitoring and early warning method based on the Internet of Things, wherein the trusted pest pressure is obtained, specifically: The pest data, the tobacco leaf damage data and the terminal state data of the continuous collection time points in the tobacco field pest situation event sequence are read according to the tobacco field numbers, the target pest quantity difference of the adjacent collection time points is calculated as the pest change quantity, and the damage area difference of the adjacent collection time points is calculated as the tobacco leaf damage change quantity; The pest change quantity corresponding to the collection time point is taken as the starting point, the tobacco leaf damage change quantity is sequentially moved backward according to the collection period, the response consistency degree of the pest change quantity and the tobacco leaf damage change quantity under each moving time interval is calculated, and the time interval corresponding to the maximum response consistency degree is screened as the response lag period; The pest change quantity and the tobacco leaf damage change quantity are matched item by item according to the response lag period, and the time sequence response verification is performed in combination with the terminal state data of the corresponding collection time point, the data with the normal terminal state and the same change direction of the pest change quantity and the tobacco leaf damage change quantity is retained, and the data with the abnormal terminal state and the different change direction of the pest change quantity and the tobacco leaf damage change quantity is marked as abnormal pest data; The abnormal pest data is used for interpolation correction with the adjacent pest data before and after the abnormal pest data and passing the time sequence response verification, and the corrected pest data is continuously arranged according to the collection time point, so that the trusted pest pressure corresponding to each collection time point of each tobacco field number is obtained.

[0010] As a preferred scheme of the tobacco pest monitoring and early warning method based on the Internet of Things, wherein the propagation corrected pest pressure sequence is obtained, specifically: The boundaries of the tobacco planting area plots are analyzed, the geometric centers of the tobacco field plots and the distances between the geometric centers of any two tobacco field plots are calculated, the tobacco field plots with a distance not more than the effective migration distance of the target pest in the pest management parameter are screened, and the tobacco field adjacency relationship is established; For each pair of tobacco field plots in the tobacco field adjacency relationship, the migration direction between the geometric centers of the two tobacco field plots is calculated, the wind direction and the wind speed of the corresponding collection time point are read from the tobacco field pest situation event sequence, the consistency degree of the wind direction and the migration direction is calculated, and the migration attenuation degree corresponding to the distance of the tobacco field plot is calculated by using the migration attenuation coefficient in the pest management parameter. The consistency degree of the wind direction and the migration direction and the migration attenuation degree corresponding to the distance of the tobacco field block are normalized respectively, and then the normalized results are multiplied to obtain the migration weight corresponding to the collection time; According to the collection time from early to late, the trusted pest pressure of each tobacco field block at each collection time is taken as the recursion basis, the migration weight corresponding to the adjacent relationship of the tobacco field is used to calculate the input trusted pest pressure of the adjacent tobacco field block, and the input trusted pest pressure of the adjacent tobacco field block and the trusted pest pressure of the current tobacco field block are recursively propagated at each collection time to obtain the propagation corrected pest pressure corresponding to each collection time. The propagation corrected pest pressure corresponding to the last two adjacent collection times is read, and the difference between the propagation corrected pest pressure corresponding to the last two adjacent collection times is taken as the recursive increment of the subsequent collection period. Starting from the last collection time, the propagation corrected pest pressure is recursively propagated according to the collection period of the tobacco pest event sequence to the predicted harvesting time. The propagation corrected pest pressure less than zero is set to zero, and the propagation corrected pest pressure corresponding to each time position from the last collection time to the predicted harvesting time is obtained. The propagation corrected pest pressure corresponding to each collection time and the propagation corrected pest pressure corresponding to each time position after the last collection time are arranged in time sequence, and the propagation corrected pest pressure sequence corresponding to each tobacco field number is obtained.

[0011] As a preferred scheme of the tobacco pest monitoring and early warning method based on the Internet of Things, wherein: the remaining loss budget is calculated according to the tobacco field management resource account, and the propagation corrected pest pressure sequence is converted into a no-control loss curve according to the pest management parameters, specifically: The expected yield of the tobacco field, the minimum income preservation value, the occurred pest damage and the occurred control cost are read from the tobacco field management resource account according to the tobacco field number, the expected yield of the tobacco field is sequentially deducted from the minimum income preservation value, the occurred pest damage and the occurred control cost, and the remaining loss budget corresponding to each tobacco field number is obtained. The propagation corrected pest pressure corresponding to the last collection time and each time position after the last collection time in the propagation corrected pest pressure sequence is read according to the tobacco field number, and the pest pressure loss conversion coefficient is read from the pest management parameter. The average propagation corrected pest pressure between adjacent time positions is calculated in time sequence, and the average propagation corrected pest pressure, the time interval of adjacent time positions and the pest pressure loss conversion coefficient are multiplied to obtain the no-control loss corresponding to each time interval. The last collection time is taken as the starting point of the no-control loss accumulation, the no-control loss corresponding to each time interval is accumulated in time sequence until the predicted harvesting time, and the accumulated no-control loss corresponding to each time position is arranged in continuous sequence with the time position as the horizontal sequence to obtain the no-control loss curve corresponding to each tobacco field number.

[0012] As a preferred scheme of the tobacco pest monitoring and early warning method based on the Internet of Things, wherein: the obtained loss after prevention and treatment and net benefit of prevention and treatment are specifically: Read the no-prevention-and-treatment loss curve according to the tobacco field number, and read the prevention and treatment costs, pest pressure reduction ratios and action duration of each prevention and treatment measure from the tobacco field management resource account book; Read the end collection time point in the no-prevention-and-treatment loss curve and each time position after the end collection time point, take each time position as a candidate start time point respectively, and arrange the candidate start time points in chronological order; Take each candidate start time point as a simulation starting point of the prevention and treatment measure, reduce the propagation corrected pest pressure after the candidate start time point and within the action duration according to the pest pressure reduction ratio, convert the reduced propagation corrected pest pressure into the loss of the corresponding time interval according to the pest pressure loss conversion coefficient in the pest management parameter, and continuously accumulate the loss without prevention and treatment outside the action duration to obtain the loss after prevention and treatment of each prevention and treatment measure corresponding to each candidate start time point; Calculate the difference between the end cumulative loss of the no-prevention-and-treatment loss curve and the loss after prevention and treatment of each prevention and treatment measure corresponding to each candidate start time point, and deduct the prevention and treatment cost of the corresponding prevention and treatment measure from the difference to obtain the net benefit of prevention and treatment of each prevention and treatment measure corresponding to each candidate start time point.

[0013] As a preferred scheme of the tobacco pest monitoring and early warning method based on the Internet of Things, wherein: the economic executable window is specifically: For each prevention and treatment measure, arrange the loss after prevention and treatment and the net benefit of prevention and treatment corresponding to the candidate start time points from early to late, and screen the first candidate start time point as the economic start time point, wherein the net benefit of prevention and treatment is not less than zero, and the sum of the loss after prevention and treatment and the prevention and treatment cost of the corresponding prevention and treatment measure does not exceed the remaining loss budget; From the economic start time point, continuously screen the candidate start time points in chronological order, wherein the net benefit of prevention and treatment is not less than zero, and the sum of the loss after prevention and treatment and the prevention and treatment cost of the corresponding prevention and treatment measure does not exceed the remaining loss budget, and take the latest candidate start time point continuously screened as the economic end time point, and the economic executable window corresponding to each prevention and treatment measure is delimited in the time range from the economic start time point to the economic end time point.

[0014] As a preferred scheme of the tobacco pest monitoring and early warning method based on the Internet of Things, wherein: the tobacco pest early warning and prevention and treatment priority order is specifically: read the economic executable window corresponding to each control measure respectively, and read the corresponding control resources from the tobacco field management resource account, take the economic opening time as the starting point, judge whether the control resources meet the required control resources of the corresponding control measure according to the time sequence, and take the first time that meets as the resource reachable time; calculate the time difference between the economic closing time and the resource reachable time corresponding to each control measure respectively, obtain the disposal margin corresponding to each control measure, and take the maximum value of the disposal margin corresponding to each control measure under the same tobacco field number as the disposal margin of the corresponding tobacco field number; generate a tobacco pest warning when the disposal margin corresponding to the tobacco field number is less than zero, arrange the tobacco field numbers according to the disposal margin corresponding to each tobacco field number from small to large to obtain the control priority order.

[0015] In a second aspect, the present application provides a tobacco pest monitoring and early warning system based on the Internet of Things, comprising, The construction module divides the tobacco field plots and numbers, and establishes a tobacco field management resource account after collecting the tobacco planting area plot boundary, pest management parameters, tobacco field management data, control measures and control resources; The alignment module collects pest data, tobacco damage data, environmental data and terminal state data according to the tobacco field number and aligns them to generate a tobacco pest event sequence; The verification module performs time sequence response verification on the tobacco pest event sequence to obtain a credible pest pressure; The recursive module identifies the adjacency relationship of the tobacco field, and recursively calculates the credible pest pressure by combining the pest management parameters and the tobacco pest event sequence after calculating the migration weight, to obtain a propagation corrected pest pressure sequence; The calculation module calculates the remaining loss budget according to the tobacco field management resource account, and converts the propagation corrected pest pressure sequence into a no-control loss curve according to the pest management parameters; The simulation module takes the no-control loss curve as the benchmark, reads the control measures and control costs from the tobacco field management resource account, and simulates the control measures at the candidate starting time to obtain the post-control loss and control net income; The window delineation module filters the candidate starting time that meets the condition that the control net income is not less than zero and the sum of the post-control loss and the control cost of the corresponding control measure does not exceed the remaining loss budget from the candidate starting time according to the remaining loss budget, the post-control loss and the control net income, and delineates the economic executable window based on the filtered candidate starting time; The result module filters the resource reachable time and calculates the disposal margin after matching the control resources with the economic executable window, and obtains the tobacco pest warning and the control priority order.

[0016] The present application has the beneficial effects that: by identifying the response lag period and calculating the credible pest pressure, the time sequence dislocation between pest invasion and tobacco damage is eliminated, the false positive rate is reduced, further by identifying the adjacency relationship of tobacco fields, combining the wind direction consistency degree and distance attenuation degree to construct the migration weight and recursively calculate the credible pest pressure, the propagation corrected pest pressure sequence is obtained, the dynamic quantitative deduction of the spatial propagation situation of the pest is realized, on this basis, the pest pressure is converted into the economic loss curve, the loss and net income after prevention and control at each candidate moment are simulated, the economic opening and closing moments are screened combined with the residual loss budget to define the economic executable window, the prevention and control opportunity is ensured to have economic efficiency and financial affordability, finally the resource reachable moment is obtained by matching the prevention and control resources and calculating the disposal margin, and the early warning and prevention and control priority are generated, the scientificity, accuracy and operability of the tobacco pest prevention and control are improved. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0018] Figure 1 The flowchart of the tobacco pest monitoring and early warning method based on the Internet of Things.

[0019] Figure 2 The space-time evolution diagram of the propagation corrected pest pressure.

[0020] Figure 3 The response lag period identification result comparison diagram.

[0021] Figure 4 The wind direction consistency degree and migration weight relationship diagram. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification.

[0023] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of the present application, therefore the present application is not limited by the specific embodiments disclosed below.

[0024] Second, the "one embodiment" or "an embodiment" referred to herein can include a particular feature, structure, or characteristic. The various embodiments are not mutually exclusive, but a single embodiment can be selected from a wider set of alternatives.

[0025] Referring to Figures 1-4 For one embodiment of the present application, the embodiment provides a tobacco pest monitoring and early warning method based on Internet of Things, comprising the following steps: S1. After collecting the tobacco planting area plot boundary, pest management parameters, tobacco field management data, prevention and control measures and prevention and control resources, the tobacco field plot is divided and numbered, and a tobacco field management resource account book is established.

[0026] S1.1: If necessary, the tobacco planting area plot boundary is collected through the geographic mapping data of the tobacco planting area. Specifically, the spatial coordinates corresponding to each boundary position in the tobacco planting area are collected by positioning collection equipment. The boundary coordinate points are connected in the order of collection to form the tobacco planting area plot boundary at the turning positions of the tobacco field boundary, the adjacent plot junction positions and the closed end point positions. According to the connection relationship between the adjacent boundary coordinate points, the closed area is identified, each closed area is divided into a tobacco field plot, and the tobacco field number is distributed according to the spatial arrangement order of the tobacco field plot to establish the corresponding relationship between the tobacco field number and the tobacco field plot.

[0027] According to the historical pest migration record, the historical pest spread range record and the pest loss statistical record of the tobacco planting area, the pest management parameters are collected. Among them, the maximum migration distance of the target pest when migrating between adjacent tobacco field plots in the historical pest migration record is read, and the maximum migration distance is taken as the effective migration distance of the target pest. The historical pest spread range record includes the spread distance corresponding to multiple historical spread events and the pest pressure value corresponding to the position of each spread distance. For example, in a historical pest spread event, the geometric center of the pest-occurring tobacco field plot is taken as the spread starting point, and the pest pressure values at the positions 、 、 、 、 distance from the spread starting point are recorded respectively . The pest pressure corresponding to the position with a distance of 0 from the spread starting point is taken as the initial pest pressure. For each spread distance position, the difference between the initial pest pressure and the pest pressure at the corresponding distance position is calculated, and the difference is sequentially operated with the initial pest pressure and the corresponding spread distance to obtain the migration attenuation coefficient of the corresponding distance position. The calculation formula is: ​

[0028] in, This represents the migration attenuation coefficient corresponding to the distance from the location. Indicates the first The pest pressure corresponding to each distance location. This represents the initial pest pressure corresponding to the point where pest spread begins. Indicates the first The diffusion distance corresponding to each distance position.

[0029] Historical pest loss statistics are retrieved from historical statistical period data. Pest pressure data and tobacco loss data for each historical statistical period are arranged from earliest to latest. Pest pressure data represents the degree of pest pressure experienced by the tobacco field within the corresponding historical statistical period, while tobacco loss data represents the economic loss of tobacco caused by pests within the corresponding historical statistical period. Following the arranged historical statistical period order, the changes in pest pressure and tobacco loss between adjacent historical statistical periods are calculated sequentially. Specifically, the pest pressure value of the previous historical statistical period is used as a benchmark; the difference between the pest pressure value of the subsequent historical statistical period and the pest pressure value of the previous historical statistical period is calculated to obtain the change in pest pressure between corresponding historical statistical periods. Simultaneously, the tobacco loss value of the subsequent historical statistical period is compared with the previous historical statistical period... The difference between tobacco leaf loss values ​​corresponding to a historical statistical period is taken as the change in tobacco leaf loss. For each group of adjacent historical statistical periods, the ratio between the change in tobacco leaf loss and the change in pest pressure is calculated to obtain the pest pressure loss conversion ratio for each historical statistical period. The pest pressure loss conversion ratio represents the degree of change in tobacco leaf loss corresponding to a unit change in pest pressure. Historical statistical period data with zero change in pest pressure are further filtered out to avoid the inability to calculate the pest pressure loss conversion ratio due to no change in pest pressure. The pest pressure loss conversion ratios corresponding to the remaining historical statistical periods are accumulated and averaged according to the number of historical statistical periods involved in the calculation. The average calculation result is taken as the pest pressure loss conversion coefficient, which is used to convert the spread-corrected pest pressure into no control loss.

[0030] S1.2: If applicable, collect the tobacco field planting area, the tobacco leaf expected unit yield, and the tobacco leaf expected purchase price. The tobacco field planting area is obtained by reading the field area record after the tobacco field is divided into plots. The tobacco leaf expected unit yield is obtained by reading the unit area yield record of the corresponding variety in the tobacco planting area, the historical harvesting record, and the current planting conditions. The tobacco leaf expected purchase price is obtained by reading the expected purchase price of the corresponding tobacco leaf grade in the tobacco purchase price record. The tobacco field expected yield is obtained by multiplying the tobacco field planting area and the tobacco leaf expected unit yield. The tobacco field expected output value is obtained by multiplying the tobacco field expected yield and the tobacco leaf expected purchase price. The income reservation amount corresponding to the minimum income requirement is obtained by multiplying the tobacco field expected output value and the minimum income proportion in the tobacco planting management requirement record. The income reservation amount is taken as the minimum income guarantee value. The pest occurrence record is read according to the tobacco field number. The damaged area and damaged area information in each pest occurrence record corresponding to the current tobacco field number are extracted. The tobacco leaf unit value of the damaged area corresponding to the current tobacco field number is obtained by matching the damaged area information with the corresponding tobacco leaf unit area value record. The tobacco leaf loss amount corresponding to each pest occurrence is obtained by multiplying the damaged area in each pest occurrence record and the corresponding tobacco leaf unit area value. The occurred pest loss corresponding to the current tobacco field number is obtained by adding the tobacco leaf loss amount corresponding to each pest occurrence according to the tobacco field number. The executed prevention and control measure type, the prevention and control frequency, and the single prevention and control cost corresponding to the current tobacco field number are obtained by reading the prevention and control record according to the tobacco field number. The single prevention and control cost corresponding to each prevention and control measure type is determined by matching the prevention and control measure type with the prevention and control measure cost record. The executed prevention and control cost corresponding to each prevention and control measure type is obtained by multiplying the prevention and control frequency and the corresponding single prevention and control cost. The occurred prevention and control cost corresponding to the current tobacco field number is obtained by adding the executed prevention and control cost corresponding to each prevention and control measure type according to the current tobacco field number. The material use cost, the personnel operation cost, and the equipment use cost corresponding to different prevention and control measure types are obtained by reading the prevention and control measure cost record according to the prevention and control measure type. The single execution cost corresponding to each prevention and control measure type is obtained by adding the material use cost, the personnel operation cost, and the equipment use cost corresponding to the same prevention and control measure type. The single execution cost corresponding to each prevention and control measure type is recorded according to the prevention and control measure type, thereby forming the prevention and control cost corresponding to each prevention and control measure.According to the tobacco production and operation management data, the expected harvesting time of each tobacco field number corresponding to the tobacco field plot is read according to the tobacco field number, and the expected harvesting time is associated with the corresponding tobacco field number. The expected harvesting time is used to indicate the time position of the tobacco field plot corresponding to the tobacco field number for expected tobacco harvesting, and serves as the time endpoint of the propagation correction pest pressure recursion, loss accumulation without control, and control resource matching after the end collection time. According to the control resource management record, the control execution plan record and the resource scheduling record, the control resources are collected, wherein the control resource management record is used to record the inventory quantity of materials, the number of personnel and the equipment configuration corresponding to each control measure, the control execution plan record is used to record the material demand, the personnel working hour demand and the equipment use time corresponding to each control measure, and the resource scheduling record is used to record the available quantity of materials, the available working hours of personnel and the available time period of equipment within the current time range; The material demand corresponding to each control measure is matched with the available quantity of materials, the personnel working hour demand is matched with the available working hours of personnel, and the equipment use time is matched with the available time period of equipment, forming a control resource including the material demand and the available quantity of materials, the personnel working hour demand and the available working hours of personnel, and the equipment use time and the available time period of equipment corresponding to each control measure.

[0031] The tobacco field number, the tobacco field plot, the pest management parameter, the tobacco field management data, the control measure and the control resource are associated, and the tobacco field plot corresponding to the same tobacco field number, the target pest effective migration distance, the migration attenuation coefficient, the pest pressure loss conversion coefficient, the tobacco field expected yield value, the minimum income preservation value, the occurred pest damage loss, the occurred control cost, the control cost corresponding to each control measure, the expected harvesting time, the control measure and the control resource are stored correspondingly, and a tobacco field management resource ledger is established.

[0032] Further, the historical pest migration record, the historical pest spread range record, the pest damage loss statistical record, the plot area record, the unit area yield record, the tobacco purchase price record, the tobacco planting management requirement record, the pest occurrence record, the control record, the control measure cost record, the control execution plan record, the resource scheduling record and the control resource management record are formed by collecting, sorting, and associating the historical planting management archives of the tobacco planting area, the Internet of Things monitoring terminal collection data, the tobacco production and operation management data, and the manually entered historical operation data.

[0033] S2: Collect pest data, tobacco leaf damage data, environmental data and terminal state data according to tobacco field number and align to generate tobacco field pest event sequence.

[0034] S2.1: It should be explained that the tobacco field number is taken as a spatial index, the data collection range of each tobacco field plot is determined according to the correspondence between the tobacco field number and the tobacco field plot, the pest monitoring terminal, the environment collection terminal and the state detection terminal are arranged in the tobacco field plot corresponding to each tobacco field number, and the data collection is synchronously performed according to the preset collection period (for example, 30 minutes); the number of target pests in the tobacco field plot is collected by the pest monitoring terminal, and the collection time corresponding to the number of target pests is recorded to form pest data; the damage situation of tobacco leaves in the tobacco field plot is collected by the image collection device or artificial inspection, the damaged area and the damaged area are identified, and the collection time corresponding to the damaged area is recorded to form tobacco leaf damage data; the wind speed and wind direction data of the corresponding position of the tobacco field plot are collected by the environment collection terminal, and the collection time corresponding to the wind speed and wind direction is recorded to form environment data; the power supply state, communication state and collection state of the pest monitoring terminal, environment collection terminal and image collection device are read by the terminal state detection terminal, and the collection time corresponding to the power supply state, communication state and collection state is recorded to form terminal state data; the pest data, tobacco leaf damage data, environment data and terminal state data corresponding to the same tobacco field number are associated according to the collection time to form the synchronous collection data record corresponding to each tobacco field number.

[0035] Further, the prevention and control records are read according to the prevention and control measure type, the tobacco field number, the prevention and control measure type and the prevention and control measure execution time corresponding to each historical prevention and control are collected, and the prevention and control measure execution data collected by the Internet of Things monitoring terminal are matched according to the tobacco field number and the prevention and control measure execution time, the pest pressure closest to the prevention and control measure execution time before the execution of the prevention and control measure and the continuous pest pressure after the execution of the prevention and control measure are read; the pest pressure before the execution of the prevention and control measure is taken as the pest pressure before the prevention and control, the lowest pest pressure during the continuous monitoring after the execution of the prevention and control measure is screened, the difference between the pest pressure before the prevention and control and the lowest pest pressure is calculated, and the difference is calculated by ratio with the pest pressure before the prevention and control to obtain the pest pressure reduction ratio of the corresponding historical prevention and control record; the pest pressure after the execution of the prevention and control measure is read from early to late according to the collection time since the prevention and control measure execution time, the collection time when the pest pressure first recovers to the pest pressure before the prevention and control is screened, and the time interval between the prevention and control measure execution time and the corresponding collection time is calculated to obtain the action duration of the corresponding historical prevention and control record; the historical prevention and control records corresponding to the same prevention and control measure are collected according to the prevention and control measure type, the pest pressure reduction ratio is calculated by average, the pest pressure reduction ratio corresponding to each prevention and control measure is obtained, the action duration is calculated by average, and the action duration corresponding to each prevention and control measure is obtained; the pest pressure reduction ratio and the action duration corresponding to each prevention and control measure are associated with the corresponding prevention and control measure, and written into the tobacco field management resource account book.

[0036] S2.2: If applicable, read the pest data, tobacco damage data, environmental data and terminal state data collected in the corresponding tobacco field plot according to the tobacco field number, extract the collection time in the pest data, tobacco damage data, environmental data and terminal state data, and sort them according to the collection time as the time reference; match the pest data, tobacco damage data, environmental data and terminal state data with the same collection time to the same time position; read the preset collection period, and take half of the preset collection period as the time matching tolerance; for data with different collection times, calculate the collection time difference, select data with a collection time difference not exceeding the time matching tolerance, and select the data with the smallest collection time difference to match to the same time position; each piece of pest data, tobacco damage data, environmental data and terminal state data is matched once and no longer participates in other time position matching; when the collection time difference between any type of data and the data corresponding to the current time position exceeds the time matching tolerance, no matching is performed; only time positions where pest data, tobacco damage data, environmental data and terminal state data are all matched are retained, and then arranged continuously from early to late according to the collection time, to generate a tobacco field pest situation event sequence corresponding to each tobacco field number.

[0037] S3: Time sequence response verification is performed on the tobacco field pest situation event sequence to obtain a credible pest pressure.

[0038] S3.1: If applicable, read the pest data, tobacco damage data and terminal state data corresponding to adjacent time positions in the tobacco field pest situation event sequence according to the tobacco field number, and extract the target pest quantity and damage area data corresponding to the two adjacent collection times, respectively; take the target pest quantity corresponding to the previous collection time as the reference, calculate the difference between the target pest quantity corresponding to the next collection time and the target pest quantity corresponding to the previous collection time, to obtain the pest change quantity corresponding to the adjacent collection times; take the damage area corresponding to the previous collection time as the reference, calculate the difference between the damage area corresponding to the next collection time and the damage area corresponding to the previous collection time, to obtain the tobacco damage change quantity corresponding to the adjacent collection times.

[0039] Read the pest change amount and the tobacco damage change amount corresponding to the continuous collection time in the tobacco field pest event sequence according to the tobacco field number, count the number of pest change amounts and tobacco damage change amounts that can participate in matching without time shift, obtain the reference matching number, take one half of the reference matching number and round up to obtain the minimum effective matching number; take the collection time corresponding to the pest change amount as the time starting point, take zero collection period as the initial shift time interval, increase the shift time interval by one collection period and move the tobacco damage change amount backward in turn, respectively count the number of participating matches that can form a corresponding relationship with the pest change amount under each shift time interval, stop increasing the shift time interval when the number of participating matches is less than the minimum effective matching number for the first time, take each shift time interval with the number of participating matches not less than the minimum effective matching number as the candidate shift time interval; for each candidate shift time interval, match the pest change amount and the tobacco damage change amount at the corresponding time position; when any one of the pest change amount and the tobacco damage change amount is equal to zero, the corresponding matching data is excluded from the response consistency calculation; for the remaining matching data, when the pest change amount and the tobacco damage change amount are both positive or both negative, record as consistent change direction, when one of the pest change amount and the tobacco damage change amount is positive and the other is negative, record as inconsistent change direction; count the number of matching data with consistent change direction, and count the number of remaining matching data after excluding zero value matching data, calculate the ratio of the number of matching data with consistent change direction and the number of remaining matching data to obtain the response consistency of the corresponding candidate shift time interval; when there is no remaining matching data after excluding zero value matching data, set the response consistency of the corresponding candidate shift time interval to zero; compare the response consistencies corresponding to each candidate shift time interval, and select the candidate shift time interval with the maximum response consistency as the response lag period, when the maximum response consistencies corresponding to two or more candidate shift time intervals are the same, select the candidate shift time interval with the minimum shift time interval as the response lag period.

[0040] As shown in Figure 3 With the increase of the candidate response lag period from 0h to 12h, the response consistency corresponding to the YT01 tobacco field fluctuates, and the response consistency reaches a maximum value of about 0.597 when the candidate response lag period is 6h, which is higher than the response consistencies corresponding to the remaining candidate response lag periods, so 6h is selected as the response lag period of the YT01 tobacco field. By sequentially shifting the tobacco damage change amount according to the collection period, and calculating the response consistency of the pest change amount and the tobacco damage change amount under each candidate response lag period, the present application can identify the time lag relationship between the change of the target pest quantity and the change of the tobacco damage, and provide a time corresponding basis for subsequent matching of the pest change amount and the tobacco damage change amount according to the response lag period and time sequence response verification.

[0041] S3.2: If applicable, read the pest damage change amount and the tobacco damage change amount matched according to the corresponding response lag period in the tobacco field pest situation event sequence according to the tobacco field number, wherein the original collection time corresponding to the pest damage change amount is taken as the matched time position, the tobacco damage change amount is moved backward according to the response lag period to form a corresponding relationship with the pest damage change amount, and the terminal state data at the original collection time corresponding to the pest damage change amount is read; the change directions of the pest damage change amount and the tobacco damage change amount are judged item by item, when the pest damage change amount and the tobacco damage change amount are both positive or both negative, it is determined that the change directions of the pest damage change amount and the tobacco damage change amount are consistent; the power supply state, the communication state and the collection state in the terminal state data at the original collection time corresponding to the pest damage change amount are read at the same time, when the power supply state, the communication state and the collection state are all in normal state, it is determined that the terminal state at the original collection time corresponding to the pest damage change amount is normal; the tobacco field pest situation event data with the terminal state normal and the change directions of the pest damage change amount and the tobacco damage change amount consistent are reserved, and the data with the terminal state abnormal and the change directions of the pest damage change amount and the tobacco damage change amount inconsistent are marked as abnormal pest data.

[0042] S3.3: If applicable, read the collection time marked as abnormal pest data according to the tobacco field number, and determine whether there is pest data passing the time sequence response verification before and after the abnormal pest data according to the collection time from early to late; when there is pest data passing the time sequence response verification before and after the single abnormal pest data, respectively extract the pest data passing the time sequence response verification closest to the front and closest to the back of the abnormal pest data, calculate the time proportion of the collection time corresponding to the abnormal pest data between the collection time corresponding to the two pest data, and calculate the difference between the two pest data according to the time proportion, obtain the pest data correction value of the collection time corresponding to the abnormal pest data; when the data corresponding to multiple consecutive collection times are marked as abnormal pest data and there is pest data passing the time sequence response verification before and after the continuous abnormal data, extract the pest data passing the time sequence response verification closest to the front and closest to the back of the continuous abnormal data, respectively calculate the time proportion of the collection time corresponding to each abnormal pest data between the collection time corresponding to the two pest data, and perform interpolation calculation according to each time proportion to obtain the pest data correction value of the collection time corresponding to each abnormal pest data; when the abnormal pest data is located at the beginning of the tobacco field pest situation event sequence and only the rear side has pest data passing the time sequence response verification, the pest data passing the time sequence response verification closest to the rear side is taken as the pest data correction value corresponding to the abnormal collection time; when the abnormal pest data is located at the end of the tobacco field pest situation event sequence and only the front side has pest data passing the time sequence response verification, the pest data passing the time sequence response verification closest to the front side is taken as the pest data correction value corresponding to the abnormal collection time; rearrange each pest data correction value and pest data passing the time sequence response verification according to the collection time from early to late to form a continuous time sequence, and take the pest data corresponding to each collection time after rearrangement as the reliable pest pressure corresponding to each tobacco field number.

[0043] S4: Identify the adjacency relationship of tobacco fields, and calculate the migration weight after combining the pest management parameters and the tobacco field pest situation event sequence to recursively the reliable pest pressure, obtain the propagation corrected pest pressure sequence.

[0044] S4.1: If applicable, read the tobacco field boundary data in the tobacco field management resource account according to the tobacco field number, extract the boundary coordinate points corresponding to each tobacco field plot; calculate the average value of the horizontal and vertical coordinates of each boundary coordinate point in the closed area formed by the continuous boundary coordinate points in the same tobacco field plot, and take the position corresponding to the average value as the geometric center of the tobacco field plot; calculate the spatial distance between the geometric centers of any two tobacco field plots respectively to obtain the distance relationship between the tobacco field plots; read the effective migration distance of the target pest in the pest management parameter, compare the spatial distance between the tobacco field plots with the effective migration distance of the target pest, screen the combination of tobacco field plots with a spatial distance less than or equal to the effective migration distance of the target pest, and establish a corresponding connection relationship for the screened combination of tobacco field plots to form the tobacco field adjacency relationship.

[0045] For each pair of tobacco fields in the tobacco field adjacency relationship, the geometric center coordinates of the first and second tobacco fields are read, the horizontal coordinate of the geometric center of the second tobacco field is subtracted from the horizontal coordinate of the geometric center of the first tobacco field to obtain a horizontal position difference value corresponding to the direction in which the first tobacco field points to the second tobacco field, the vertical coordinate of the geometric center of the second tobacco field is subtracted from the vertical coordinate of the geometric center of the first tobacco field to obtain a vertical position difference value corresponding to the direction in which the first tobacco field points to the second tobacco field, the arctangent of the horizontal position difference value and the vertical position difference value is calculated to obtain the migration direction angle of the first tobacco field pointing to the second tobacco field; the horizontal coordinate of the geometric center of the first tobacco field is subtracted from the horizontal coordinate of the geometric center of the second tobacco field to obtain a horizontal position difference value corresponding to the direction in which the second tobacco field points to the first tobacco field, the vertical coordinate of the geometric center of the first tobacco field is subtracted from the vertical coordinate of the geometric center of the second tobacco field to obtain a vertical position difference value corresponding to the direction in which the second tobacco field points to the first tobacco field, the arctangent of the horizontal position difference value and the vertical position difference value is calculated to obtain the migration direction angle of the second tobacco field pointing to the first tobacco field. The wind direction and wind speed at the corresponding collection time are read from the tobacco field pest event sequence, the wind direction is converted into a wind direction angle according to a preset wind direction angle correspondence relationship, wherein the north direction is taken as the 0° reference, the east wind, south wind and west wind are converted into 90°, 180° and 270° respectively in the clockwise direction, and the northeast wind, southeast wind, southwest wind and northwest wind are converted into 45°, 135°, 225° and 315° respectively to obtain the wind direction angle at the corresponding collection time; the absolute difference between the wind direction angle and the migration direction angle of the first tobacco field pointing to the second tobacco field, and the absolute difference between the wind direction angle and the migration direction angle of the second tobacco field pointing to the first tobacco field are calculated respectively, when the absolute difference is not more than 180°, the absolute difference is taken as the angle difference value corresponding to the migration direction, when the absolute difference exceeds 180°, 360° is subtracted from the absolute difference to obtain the angle difference value corresponding to the migration direction; the angle difference value corresponding to the two migration directions is respectively compared with 180 to obtain the angle difference proportion corresponding to the migration direction, the angle difference proportion corresponding to the migration direction is subtracted from the value 1 respectively to obtain the consistency degree of the wind direction and the migration direction corresponding to the first tobacco field pointing to the second tobacco field, and the consistency degree of the wind direction and the migration direction corresponding to the second tobacco field pointing to the first tobacco field, wherein the consistency degree of the wind direction and the migration direction is used to represent the promotion degree of the wind direction at the corresponding collection time to the migration direction of the pests between the tobacco fields. The migration attenuation coefficient in the pest management parameter is read, the distance between the geometric centers of the two tobacco fields is calculated, and the distance between the geometric centers of the two tobacco fields is multiplied by the migration attenuation coefficient to obtain a migration attenuation index corresponding to the distance; the migration attenuation index is subtracted from the value 1 to obtain the migration attenuation degree corresponding to the distance between the tobacco fields; when the migration attenuation degree is less than zero, the migration attenuation degree is corrected to zero.

[0046] Further, the historical wind direction is collected to establish a preset wind direction angle corresponding relationship. Specifically, the wind direction types are divided into eight directions according to geographical directions, i.e., north, northeast, east, southeast, south, southwest, west and northwest. The north direction is taken as a 0° reference and the angle is increased in a clockwise direction. Specifically, the north corresponds to 0°, the northeast corresponds to 45°, the east corresponds to 90°, the southeast corresponds to 135°, the south corresponds to 180°, the southwest corresponds to 225°, the west corresponds to 270°, and the northwest corresponds to 315°, thereby forming the preset wind direction angle corresponding relationship between the wind direction types and the wind direction angles.

[0047] S4.2: If necessary, for each collection time, the consistency degree of the wind direction and the migration direction corresponding to two migration directions in the whole tobacco planting area is collected, the maximum wind direction consistency degree and the minimum wind direction consistency degree are screened out, and the consistency degree of the wind direction and the migration direction corresponding to each migration direction is normalized by the minimum-maximum normalization method; the migration attenuation degree corresponding to the tobacco field plot distance corresponding to the whole tobacco field adjacency relationship is collected, the maximum migration attenuation degree and the minimum migration attenuation degree are screened out, and the migration attenuation degree corresponding to each tobacco field adjacency relationship is normalized by the minimum-maximum normalization method; when the maximum value and the minimum value of the corresponding index are not equal, the current value is subtracted from the minimum value, and the difference between the maximum value and the minimum value is divided to obtain the corresponding normalized result; when the maximum value and the minimum value of the corresponding index are equal and both are zero, the corresponding normalized result is set to zero; when the maximum value and the minimum value of the corresponding index are equal and both are greater than zero, the corresponding normalized result is set to one; the wind direction consistency degree normalization result corresponding to each migration direction and the migration attenuation degree normalization result corresponding to the corresponding tobacco field adjacency relationship are multiplied to obtain the migration weight corresponding to each migration direction at the corresponding collection time.

[0048] As Figure 4As shown, when the distance of the tobacco field plot remains unchanged, the migration weight continuously increases with the increase of the consistency degree of the wind direction and the migration direction; when the consistency degree of the wind direction and the migration direction is the same, the migration weight decreases in turn when the distance of the tobacco field plot increases from 150 m to 550 m. When the consistency degree of the wind direction and the migration direction is 1.0, the migration weights corresponding to the distances of 150 m, 300 m, 450 m and 550 m of the tobacco field plot are about 0.75, 0.50, 0.25 and 0.08 respectively; in the locally enlarged low consistency degree interval, the migration weights corresponding to different distances of the tobacco field plot still maintain obvious differences, when the consistency degree of the wind direction and the migration direction is about 0.34, the migration weight corresponding to the distance of 150 m of the tobacco field plot is about 0.255, and the difference between the migration weight corresponding to the distance of 550 m of the tobacco field plot is about 0.227; it should be noted that the plurality of curves correspond to the migration weights under different distances of the tobacco field plot. The results show that the migration weight can reflect the wind direction promoting effect and the distance attenuation effect of the tobacco field plot by considering the consistency degree of the wind direction and the migration direction and the migration attenuation degree corresponding to the distance of the tobacco field plot, thereby providing a quantitative basis for calculating the credible pest pressure input by the adjacent tobacco field plot.

[0049] S4.3: It should be noted that the credible pest pressure of each tobacco field plot at each collection time is read according to the tobacco field number, and the collection time is sorted from early to late, and the credible pest pressure of the current tobacco field plot at the collection time is taken as the basic pest pressure of the current tobacco field plot at the collection time; for each adjacent tobacco field plot in the tobacco field adjacency relationship of the current tobacco field plot, the migration weight corresponding to the migration direction from the adjacent tobacco field plot to the current tobacco field plot at the corresponding collection time is read, and the credible pest pressure of the adjacent tobacco field plot at the corresponding collection time is multiplied by the corresponding migration weight to obtain the credible pest pressure input by the adjacent tobacco field plot; the basic pest pressure of the current tobacco field plot at the collection time is added to the credible pest pressure input by all adjacent tobacco field plots to obtain the propagation corrected pest pressure corresponding to the current collection time.

[0050] The propagation corrected pest pressure corresponding to the end collection time is read according to the tobacco field number, the propagation corrected pest pressure corresponding to the later collection time is subtracted from the propagation corrected pest pressure corresponding to the earlier collection time, and the propagation corrected pest pressure change amount corresponding to the end collection period is obtained; the propagation corrected pest pressure change amount is taken as the recursive increment of the subsequent collection period, the propagation corrected pest pressure corresponding to the end collection time is taken as the recursive starting point, the propagation corrected pest pressure change amount is superimposed period by period according to the collection period corresponding to the tobacco pest event sequence, and the propagation corrected pest pressure corresponding to the next collection period is obtained; the period-by-period recursive process is repeatedly executed until the expected harvest time is reached; the propagation corrected pest pressure corresponding to each recursive period is judged, and when the propagation corrected pest pressure is less than zero, the propagation corrected pest pressure is adjusted to zero, the recursive result of the propagation corrected pest pressure not less than zero is retained, and the propagation corrected pest pressure corresponding to each time position from the end collection time to the expected harvest time is obtained.

[0051] The propagation corrected pest pressure corresponding to each collection time is read according to the tobacco field number, the propagation corrected pest pressure corresponding to each collection time before the end collection time is extracted, and the propagation corrected pest pressure is sorted according to the collection time from early to late; the propagation corrected pest pressure corresponding to each time position from the end collection time to the expected harvest time is read, the propagation corrected pest pressure corresponding to each collection time before the end collection time is spliced and arranged according to the time sequence with the propagation corrected pest pressure corresponding to each time position after the end collection time, and a continuous time change sequence is formed; the arranged time positions and the corresponding propagation corrected pest pressure are associated and recorded, and the propagation corrected pest pressure sequence corresponding to each tobacco field number is obtained.

[0052] As shown in Figure 2 , the propagation corrected pest pressure corresponding to each tobacco field number presents different time sequence evolution characteristics with cumulative time, and there is a significant difference in the propagation corrected pest pressure corresponding to different tobacco field numbers at the same cumulative time position, and the propagation corrected pest pressure also changes continuously with cumulative time in the same tobacco field number; among them, the middle part of the tobacco field number forms a higher propagation corrected pest pressure area in the range of about 1.4x10<2>h to 2.0x10<2>h, and the highest propagation corrected pest pressure exceeds 120, while some tobacco field numbers still maintain a lower propagation corrected pest pressure in the same time range. The results show that the present application can calculate the reliable pest pressure input by the adjacent tobacco field blocks through the adjacent relationship of the tobacco field and the migration weight, and continuously calculate the reliable pest pressure of the current tobacco field block according to the collection time, which can represent the differential propagation change of the pest pressure between different tobacco field blocks with time, and provide continuous spatio-temporal pest pressure data basis for converting the propagation corrected pest pressure sequence into a no-control loss curve.

[0053] S5: Calculate the remaining loss budget according to the tobacco field management resource account book, and convert the spread corrected pest pressure sequence into the no-control loss curve according to the pest management parameters.

[0054] S5.1: If applicable, read the corresponding tobacco field expected yield, minimum income maintenance value, pest damage loss occurred, and control cost occurred in the tobacco field management resource account book according to the tobacco field number. Take the tobacco field expected yield as the initial amount and subtract the minimum income maintenance value to obtain the disposable loss amount after deducting the minimum income requirement. Subtract the disposable loss amount from the pest damage loss occurred to obtain the remaining amount after deducting the current pest impact. Continue to deduct the control cost occurred from the remaining amount to obtain the remaining loss budget corresponding to the current tobacco field number. Record the remaining loss budget corresponding to each tobacco field number according to the tobacco field number.

[0055] S5.2: If applicable, read the end collection time and the spread corrected pest pressure corresponding to each time position after the end collection time in the spread corrected pest pressure sequence according to the tobacco field number, and arrange the spread corrected pest pressure corresponding to each time position in chronological order. Read the pest pressure loss conversion coefficient from the pest management parameters, and sequentially obtain the spread corrected pest pressure corresponding to adjacent time positions. Add the spread corrected pest pressure corresponding to the latter time position to the spread corrected pest pressure corresponding to the former time position, and then perform the quotient operation with the value 2 to obtain the average spread corrected pest pressure between adjacent time positions. Calculate the time interval between adjacent time positions, and multiply the average spread corrected pest pressure, the time interval of adjacent time positions, and the pest pressure loss conversion coefficient to obtain the no-control loss caused by the spread corrected pest pressure in the corresponding time interval. Calculate the corresponding no-control loss between each adjacent time position in chronological order to form the no-control loss corresponding to each time interval.

[0056] According to the no-control loss corresponding to each time interval, take the end collection time as the starting point of no-control loss accumulation, and set the cumulative no-control loss corresponding to the end collection time to zero. Sequentially extract the no-control loss corresponding to each time interval after the end collection time in chronological order, and add the no-control loss corresponding to the current time interval to the cumulative no-control loss corresponding to the previous time position to obtain the cumulative no-control loss corresponding to the current time position. Repeat the no-control loss accumulation process corresponding to the time interval until the predicted harvesting time is reached. Arrange the cumulative no-control loss corresponding to each time position from early to late, and record the time position and the cumulative no-control loss corresponding to each time position to form the no-control loss curve corresponding to each tobacco field number.

[0057] S6: Taking the no-control loss curve as a benchmark, read the control measures and control costs from the tobacco field management resource account, simulate the control measures at the candidate start time, and obtain the post-control loss and control net income.

[0058] S6.1: If applicable, read the no-control loss curve according to the tobacco field number, and read the control costs, pest pressure reduction ratio, and action duration of each control measure from the tobacco field management resource account. Read the end collection time and the cumulative no-control loss corresponding to each time position after the end collection time to the expected harvest time in the no-control loss curve. Extract the end collection time and each time position after the end collection time as the candidate start time, and arrange the candidate start times in chronological order to form a candidate start time sequence for subsequent control measure action time matching.

[0059] Take each candidate start time as the simulation starting point of the corresponding control measure, read the pest pressure reduction ratio and action duration of the corresponding control measure from the tobacco field management resource account. Traverse the propagation corrected pest pressure sequence backward from the candidate start time, and select the propagation corrected pest pressure corresponding to each time position after the candidate start time and within the action duration. Multiply the difference between 1 and the pest pressure reduction ratio by the propagation corrected pest pressure corresponding to each time position to obtain the corrected propagation corrected pest pressure after the corresponding control measure is executed. Average the corrected propagation corrected pest pressures corresponding to adjacent time positions to obtain the post-control average pest pressure. Read the pest pressure loss conversion coefficient in the pest management parameter, and multiply the post-control average pest pressure, the time interval between adjacent time positions, and the pest pressure loss conversion coefficient to obtain the post-control loss corresponding to each time interval within the action duration. Replace the no-control loss in the corresponding time interval with the post-control loss corresponding to each time interval within the action duration, and accumulate the no-control loss corresponding to each time position outside the action duration in chronological order to obtain the post-control loss produced by the corresponding control measure at the corresponding candidate start time.

[0060] S6.2: If applicable, read the cumulative loss without control corresponding to the end collection time of the loss without control curve and the predicted harvest time according to the tobacco field number, and take the cumulative loss without control as the end cumulative loss of the loss without control curve; read the loss after control corresponding to each candidate start time of each control measure respectively, and calculate the difference between the end cumulative loss of the loss without control curve and the loss after control of the corresponding control measure at the corresponding candidate start time to obtain the loss reduction amount of the corresponding control measure at the corresponding candidate start time; read the control cost of the corresponding control measure from the tobacco field management resource account book, and calculate the difference between the loss reduction amount and the control cost of the corresponding control measure to obtain the net control benefit of each control measure at each candidate start time.

[0061] S7: According to the remaining loss budget, the loss after control and the net control benefit, screen the candidate start time that meets the condition that the net control benefit is not less than zero and the sum of the loss after control and the control cost of the corresponding control measure does not exceed the remaining loss budget from the candidate start time, and determine the economically executable window based on the screened candidate start time.

[0062] S7.1: If applicable, read the loss after control and the net control benefit of each control measure at each candidate start time according to the tobacco field number respectively, and sort the loss after control and the net control benefit corresponding to the same control measure from early to late according to the candidate start time; for each candidate start time after sorting, judge whether the corresponding net control benefit is not less than zero, and at the same time, add the corresponding loss after control and the control cost of the corresponding control measure to obtain the loss after control cost occupation value, and judge whether the loss after control cost occupation value exceeds the remaining loss budget in the tobacco field management resource account book; screen the candidate start time that meets the condition that the net control benefit is not less than zero and the loss after control cost occupation value does not exceed the remaining loss budget at the same time; when there is a candidate start time that meets the condition, select the first candidate start time that meets the condition according to the candidate start time sorting result, and determine the first candidate start time that meets the condition as the economic start time of the corresponding control measure; when there is no candidate start time that meets the condition, mark the corresponding control measure as economically unexecutable, and do not generate the economic start time, the economic shutdown time and the economic executable window of the corresponding control measure.

[0063] S7.2: If applicable, read the economic opening time corresponding to each control measure according to the tobacco field number, and identify the control measures marked as economically unexecutable; for the control measures not marked as economically unexecutable, take the candidate starting time corresponding to the economic opening time as the starting position, read the subsequent candidate starting times in the order of early to late, and screen the candidate starting times that meet the conditions according to the judgment process in S7.1; determine the latest candidate starting time obtained by continuous screening as the economic closing time, and take the time range between the economic opening time and the economic closing time as the economic executable window of the corresponding control measure; for the control measures marked as economically unexecutable, do not perform the screening of the economic closing time and the economic executable window.

[0064] S8: After matching the control resources and the economic executable window, screen the resource reachable time and calculate the disposal margin to obtain the tobacco pest warning and the control priority.

[0065] S8.1: If applicable, read the economic executable window corresponding to each control measure according to the tobacco field number, and identify the control measures marked as economically unexecutable; for the control measures with the economic executable window, read the control resources corresponding to the control measures from the tobacco field management resource account book, wherein the control resources include the material demand and the available amount of materials, the personnel working hour demand and the available working hours of personnel, the equipment use time and the available time period of equipment; take the economic opening time as the time starting point, read the control resources corresponding to each time position between the economic opening time and the expected harvesting time in the order of time, compare the available amount of materials corresponding to each time position with the material demand of the control measures, compare the available working hours of personnel with the personnel working hour demand of the control measures, and compare the available time period of equipment with the equipment use time of the control measures; when the available amount of materials is not less than the material demand, the available working hours of personnel are not less than the personnel working hour demand, and the available time period of equipment meets the equipment use time requirement, determine the corresponding time position as the time position that meets the control resource requirements; screen the time position that first meets the control resource requirements in the order of time, and determine the time position that first meets the control resource requirements as the resource reachable time; when there is no time position that meets the control resource requirements between the economic opening time and the expected harvesting time, mark the corresponding control measures as resource unreachable before the expected harvesting time.

[0066] S8.2: If applicable, the disposal margin of each control measure is obtained by subtracting the economic closing time from the resource available time corresponding to the control measure; when the resource available time is not later than the economic closing time, the disposal margin of the corresponding control measure is not less than zero, and when the resource available time is later than the economic closing time, the disposal margin of the corresponding control measure is less than zero; the disposal margins of each control measure are collected according to the tobacco field number, and the maximum disposal margin under the same tobacco field number is selected as the disposal margin of the corresponding tobacco field number; when all control measures under the same tobacco field number are marked as resource unavailable before the predicted harvesting time, a tobacco pest warning is generated for the corresponding tobacco field number, and the corresponding tobacco field number is marked as a resource completely unavailable tobacco field; a tobacco pest warning is generated when the disposal margin of the corresponding tobacco field number is less than zero; when there are multiple resource completely unavailable tobacco fields, the economic closing time corresponding to each control measure in each resource completely unavailable tobacco field is read, the earliest economic closing time under the same tobacco field number is selected, and the resource completely unavailable tobacco fields are arranged according to the earliest economic closing time corresponding to each tobacco field number from early to late; in generating the control priority order, the resource completely unavailable tobacco fields are arranged before the tobacco fields with disposal margins, and the remaining tobacco field numbers are arranged according to the disposal margins from small to large, to obtain the control priority order.

[0067] The embodiment also provides a tobacco pest monitoring and early warning system based on the Internet of Things, which comprises: The construction module collects the boundaries of tobacco planting areas, pest management parameters, tobacco field management data, control measures and control resources, divides the tobacco fields and numbers them, and establishes a tobacco field management resource account; The alignment module collects pest data, tobacco damage data, environmental data and terminal state data according to the tobacco field number and aligns them to generate a tobacco pest event sequence; The verification module performs time sequence response verification on the tobacco pest event sequence to obtain a credible pest pressure; The recursive module identifies the adjacency relationship of the tobacco fields, calculates the migration weight by combining the pest management parameters and the tobacco pest event sequence, and recursively calculates the credible pest pressure to obtain a propagation corrected pest pressure sequence; The calculation module calculates the remaining loss budget according to the tobacco field management resource account, and converts the propagation corrected pest pressure sequence into a no-control loss curve according to the pest management parameters; The simulation module takes the no-control loss curve as a reference, reads the control measures and control costs from the tobacco field management resource account, simulates the control measures at the candidate starting time, and obtains the post-control loss and control net income; The windowing module determines the economic executable window according to the remaining loss budget, the loss after prevention and treatment, and the net benefit of prevention and treatment, screens the candidate starting time that meets the conditions of the net benefit of prevention and treatment being not less than zero and the sum of the loss after prevention and treatment and the prevention and treatment cost corresponding to the prevention and treatment measure being not more than the remaining loss budget from the candidate starting time, and determines the economic executable window based on the screened candidate starting time. The result module screens the resource reachable time and calculates the disposal margin after matching the prevention and treatment resource with the economic executable window, and obtains the tobacco pest warning and prevention priority.

[0068] In summary, the present application eliminates the time sequence misalignment between pest invasion and tobacco damage by identifying the response lag period and calculating the credible pest pressure, reduces the false positive rate, further realizes the dynamic quantitative deduction of the spatial transmission situation of the pest by identifying the adjacency relationship of the tobacco field, combining the wind direction consistency degree and the distance attenuation degree to construct the migration weight and recursively calculate the credible pest pressure, obtains the transmission corrected pest pressure sequence, converts the pest pressure into the economic loss curve, simulates the loss and net benefit after prevention and treatment at each candidate time, screens the economic opening and closing time in combination with the remaining loss budget to determine the economic executable window, ensures that the prevention and treatment opportunity has economic efficiency and financial affordability, finally matches the prevention and treatment resource to obtain the resource reachable time and calculates the disposal margin, generates the warning and prevention priority, and improves the scientificity, accuracy and operability of the tobacco pest prevention and control.

[0069] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit it, although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A method for monitoring and early warning of tobacco pests based on the Internet of Things, characterized in that, include: After collecting data on tobacco planting area boundaries, pest management parameters, tobacco field management data, control measures and control resources, tobacco field plots are divided and numbered, and a tobacco field management resource ledger is established. Data on pests, tobacco leaf damage, environmental conditions, and terminal status were collected and aligned according to the tobacco field number to generate a sequence of tobacco field pest events. Time-series response verification was performed on the insect infestation event sequence in tobacco fields to obtain reliable insect infestation pressure. Identify the adjacency relationships of tobacco fields, and calculate the migration weight by combining pest management parameters and tobacco field pest event sequences to recursively infer the reliable pest pressure and obtain the propagation-corrected pest pressure sequence. The remaining loss budget was calculated based on the tobacco field management resource ledger, and the propagation modified pest pressure sequence was converted into a no-control loss curve according to pest management parameters. Based on the no-control loss curve, after reading the control measures and control costs from the tobacco field management resource ledger, the control measures are simulated at the candidate start time to obtain the post-control loss and net control benefit. Based on the remaining loss budget, post-control loss, and net control benefit, select candidate start times from the candidate start times that satisfy the conditions that the net control benefit is not less than zero and the sum of the post-control loss and the control cost of the corresponding control measures does not exceed the remaining loss budget, and define the economically feasible window based on the selected candidate start times. After matching prevention and control resources with economically feasible windows, the available time of resources is screened and the disposal margin is calculated to obtain tobacco pest early warning and prevention and control priority.

2. The method for monitoring and early warning of tobacco pests based on the Internet of Things as described in claim 1, characterized in that: The pest management parameters include the effective migration distance of the target pest, the migration attenuation coefficient, and the pest pressure loss conversion coefficient. The tobacco field management data includes the estimated output value of the tobacco field, the minimum guaranteed income value, the pest losses that have occurred, the control costs that have been incurred, the control costs corresponding to each control measure, and the estimated harvest time. The prevention and control resources include the material demand and available material quantity corresponding to each prevention and control measure, the manpower demand and available manpower hours, the equipment usage time and available equipment time period.

3. The method for monitoring and early warning of tobacco pests based on the Internet of Things as described in claim 2, characterized in that: The generated tobacco field insect infestation event sequence is specifically as follows: Using the tobacco field number as a spatial index, pest data, tobacco leaf damage data, environmental data, and terminal status data are collected synchronously in the tobacco field plot corresponding to each tobacco field number. The pest data records the number of target pests and the time of collection, the tobacco leaf damage data records the damaged area and the time of collection, the environmental data records the wind speed, wind direction, and the time of collection, and the terminal status data records the power supply status, communication status, collection status, and collection time. Data on pests, tobacco leaf damage, environmental conditions, and terminal status were collected separately according to the tobacco field number. Data from different collection times were sorted based on the collection time. Data on pests, tobacco leaf damage, environmental conditions, and terminal status collected at the same time were matched to the same time position. Data with different collection times were calculated based on the difference in collection time, and data with the smallest difference in collection time were matched to the same time position. Finally, the data were arranged continuously from earliest to latest according to the collection time to generate a sequence of tobacco field pest events corresponding to each tobacco field number.

4. The method for monitoring and early warning of tobacco pests based on the Internet of Things as described in claim 3, characterized in that: The obtained reliable pest pressure is specifically as follows: According to the tobacco field number, read the pest data, tobacco leaf damage data and terminal status data of the continuous collection time in the tobacco field pest event sequence, calculate the difference in the number of target pests at adjacent collection time as the pest change, and calculate the difference in the damaged area at adjacent collection time as the tobacco leaf damage change. Starting from the collection time corresponding to the change in pest damage, the change in tobacco leaf damage was shifted sequentially according to the collection cycle. The response consistency between the change in pest damage and the change in tobacco leaf damage was calculated at each shift time interval. The time interval corresponding to the maximum response consistency was selected as the response lag period. Match the changes in pests and tobacco damage according to the response lag period, and perform time-series response verification by combining the terminal status data at the corresponding collection time. Data with normal terminal status and the same direction of change in pests and tobacco damage are retained, while data with abnormal terminal status and data with inconsistent direction of change in pests and tobacco damage are marked as abnormal pest data. By interpolating and correcting the pest data that are adjacent to the abnormal pest data and have passed the time-series response verification, and then arranging the corrected pest data continuously according to the collection time, the credible pest pressure corresponding to each tobacco field number at each collection time is obtained.

5. The method for monitoring and early warning of tobacco pests based on the Internet of Things as described in claim 4, characterized in that: The obtained propagation-corrected pest pressure sequence is specifically as follows: Analyze the boundaries of tobacco planting areas, calculate the geometric center of each tobacco field and the distance between the geometric centers of any two tobacco fields, screen out tobacco fields whose distance does not exceed the effective migration distance of the target pest in the pest management parameters, and establish the adjacency relationship of tobacco fields. For each pair of tobacco field plots in the tobacco field adjacency relationship, the migration direction between the geometric centers of the two tobacco field plots is calculated. The wind direction and wind speed at the corresponding collection time are read from the tobacco field pest event sequence. The consistency between the wind direction and the migration direction is calculated. The migration attenuation coefficient in the pest management parameters is used to calculate the migration attenuation degree corresponding to the distance between tobacco field plots. The degree of consistency between wind direction and migration direction and the degree of migration attenuation corresponding to the distance of tobacco field plots were normalized respectively. The normalization results were then multiplied to obtain the migration weight at the corresponding collection time. According to the collection time from morning to night, the credible pest pressure corresponding to each tobacco field at each collection time is used as the basis for recursion. The migration weight corresponding to the tobacco field adjacency relationship is used to calculate the credible pest pressure input of the adjacent tobacco field. The credible pest pressure input of the adjacent tobacco field and the credible pest pressure of the current tobacco field are recursively calculated for each collection time to obtain the propagation-corrected pest pressure corresponding to each collection time. Read the propagation-corrected pest pressure corresponding to the two adjacent collection times at the end. Use the difference between the propagation-corrected pest pressure corresponding to the two adjacent collection times at the end as the incremental value for subsequent collection cycles. Starting from the end collection time, advance the collection cycle according to the tobacco field pest event sequence to the expected harvest time. Set the propagation-corrected pest pressure with an advance result less than zero to zero to obtain the propagation-corrected pest pressure corresponding to each time position from the end collection time to the expected harvest time. The propagation-corrected pest pressure corresponding to each collection time and the propagation-corrected pest pressure corresponding to each time position after the final collection time are arranged continuously in chronological order to obtain the propagation-corrected pest pressure sequence corresponding to each tobacco field number.

6. The method for monitoring and early warning of tobacco pests based on the Internet of Things as described in claim 5, characterized in that: The remaining loss budget is calculated based on the tobacco field management resource ledger, and the propagation modified pest pressure sequence is converted into a no-control loss curve according to pest management parameters. Specifically: According to the tobacco field number, the estimated output value, minimum guaranteed income value, incurred pest losses and incurred control costs of the tobacco field are read from the tobacco field management resource ledger. The minimum guaranteed income value, incurred pest losses and incurred control costs are deducted from the estimated output value of the tobacco field in sequence to obtain the remaining loss budget corresponding to each tobacco field number. According to the tobacco field number, read the transmission modified pest pressure corresponding to the end collection time and the time position after the end collection time in the transmission modified pest pressure sequence, and read the pest pressure loss conversion coefficient from the pest management parameters. Calculate the average transmission modified pest pressure between adjacent time positions according to the time sequence. Multiply the average transmission modified pest pressure, the time interval between adjacent time positions and the pest pressure loss conversion coefficient to obtain the no-control loss corresponding to each time interval. Using the end-of-pipe collection time as the starting point for the accumulation of non-control losses, the non-control losses corresponding to each time interval are accumulated in chronological order until the expected harvest time. The accumulated non-control losses corresponding to each time position are arranged continuously in chronological order to obtain the non-control loss curve corresponding to each tobacco field number.

7. The method for monitoring and early warning of tobacco pests based on the Internet of Things as described in claim 6, characterized in that: The losses and net benefits obtained after prevention and control are specifically as follows: Read the loss curve without control according to the tobacco field number, and read the control cost, pest pressure reduction ratio and duration of effect of each control measure from the tobacco field management resource ledger. Read the end collection time and subsequent time positions in the no-prevention loss curve, take each time position as a candidate start time, and arrange the candidate start times in chronological order. Using each candidate start time as the simulation starting point of the control measures, the propagation-corrected pest pressure after the candidate start time and within the duration of action is reduced according to the pest pressure reduction ratio. Then, the reduced propagation-corrected pest pressure is converted into the loss of the corresponding time interval according to the pest pressure loss conversion coefficient in the pest management parameters. This loss is continuously accumulated with the loss of no control outside the duration of action to obtain the post-control loss of each control measure at each candidate start time. Calculate the difference between the cumulative loss at the end of the no-control loss curve and the post-control loss corresponding to each control measure at the start time of each candidate, and subtract the control cost of the corresponding control measure from the difference to obtain the net control benefit corresponding to each control measure at the start time of each candidate.

8. The method for monitoring and early warning of tobacco pests based on the Internet of Things as described in claim 7, characterized in that: The definition of the economically feasible window is as follows: For each prevention and control measure, the corresponding post-control losses and net control benefits are arranged in order of candidate start time from earliest to latest. The first candidate start time that has a net control benefit of not less than zero and the sum of post-control losses and the control costs of the corresponding prevention and control measures does not exceed the remaining loss budget is selected as the economic start time. Starting from the moment the economy begins, candidate start times are continuously screened in chronological order where the net benefit of prevention and control is not less than zero and the sum of the loss after prevention and control and the prevention and control cost of the corresponding prevention and control measures does not exceed the remaining loss budget. The latest candidate start time selected in the continuous screening is taken as the moment the economy ends. The economically feasible window corresponding to each prevention and control measure is defined by the time range from the moment the economy begins to the moment the economy ends.

9. The method for monitoring and early warning of tobacco pests based on the Internet of Things as described in claim 8, characterized in that: The specific priority order for obtaining tobacco pest early warning and control is as follows: Read the economically feasible window corresponding to each prevention and control measure, and read the corresponding prevention and control resources from the tobacco field management resource ledger. Starting from the economic start time, judge whether the prevention and control resources meet the prevention and control resources required by the corresponding prevention and control measures in time sequence, and select the first time to meet the requirements as the resource reachable time. Calculate the time difference between the economic shutdown time and the resource availability time corresponding to each prevention and control measure to obtain the disposal margin corresponding to each prevention and control measure. Select the maximum value among the disposal margins corresponding to each prevention and control measure under the same tobacco field number as the disposal margin of the corresponding tobacco field number. When the treatment margin corresponding to a tobacco field number is less than zero, a tobacco pest warning is generated. The tobacco field numbers are arranged in ascending order of treatment margin to obtain the priority order for prevention and control.

10. A tobacco pest monitoring and early warning system based on the Internet of Things (IoT), used to execute the tobacco pest monitoring and early warning method based on the IoT as described in any one of claims 1 to 9, characterized in that, include: The module collects data on tobacco planting area boundaries, pest management parameters, tobacco field management data, control measures and control resources, then divides and numbers tobacco field plots, and establishes a tobacco field management resource ledger. The alignment module collects and aligns pest data, tobacco leaf damage data, environmental data, and terminal status data according to the tobacco field number, generating a tobacco field pest event sequence; The verification module performs time-series response verification on the insect pest event sequence in the tobacco field to obtain reliable pest pressure. The recursive module identifies the adjacency relationships of tobacco fields and calculates the migration weight by combining pest management parameters and tobacco field pest event sequences, and then recursively calculates the reliable pest pressure to obtain the propagation-corrected pest pressure sequence. The calculation module calculates the remaining loss budget based on the tobacco field management resource ledger and converts the propagation modified pest pressure sequence into a no-control loss curve according to pest management parameters; The simulation module, based on the loss curve without control, reads the control measures and control costs from the tobacco field management resource ledger, and then simulates the control measures at the candidate start time to obtain the loss and net benefit of control after control. The window module is defined by selecting candidate start times from the candidate start times based on the remaining loss budget, post-control loss, and net control benefit. These candidate start times satisfy the condition that the net control benefit is not less than zero and the sum of the post-control loss and the control cost of the corresponding control measures does not exceed the remaining loss budget. An economically feasible window is then defined based on the selected candidate start times. The results module matches prevention and control resources with economically feasible windows, filters out resource availability times and calculates disposal margins, and obtains tobacco pest early warning and prevention and control priorities.

Images