A crop disease model construction method and system based on multi-environment factor coupling
By constructing a crop disease model coupled with multiple environmental factors, the problem of lack of uniformity in existing disease models is solved, enabling reuse and rapid adaptation of multiple diseases, supporting platform-based deployment, and possessing high generalization ability and interpretability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- TIANJIN TIANYI TECHNOLOGY CO LTD
- Filing Date
- 2026-05-28
- Publication Date
- 2026-06-26
AI Technical Summary
The lack of uniformity in existing agricultural disease prediction models means that the entire prediction chain needs to be redeveloped for each new disease, which is costly and difficult to reuse, making it impossible to form a unified platform deployment solution.
We construct a crop disease model based on the coupling of multiple environmental factors. By acquiring environmental, soil and crop data, we calculate the spore attachment success rate and infection activity. Using an infection state machine model and multiple correction coefficients, we uniformly describe the disease process and support platform-based expansion.
It enables the reuse and rapid adaptation of multiple diseases, shortens the development cycle of new diseases, supports platform-based deployment, and has high generalization ability and interpretability.
Smart Images
Figure CN122287160A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of smart agriculture and plant protection technology, and in particular to a method and system for constructing crop disease models based on the coupling of multiple environmental factors. Background Technology
[0002] Current agricultural disease prediction models suffer from significant fragmentation. Different models are often required for different diseases such as rice blast, wheat scab, and corn rust, leading to extremely high development costs, difficulty in model reuse, and the inability to form a unified platform deployment solution. Existing technologies employ threshold models, empirical statistical models, and mechanistic models, resulting in inconsistent model structures and a lack of universal abstraction of disease biological processes. Specifically, existing technologies fail to abstract common processes such as spore production, dispersal, deposition, germination, and infection into a unified module, necessitating the redevelopment of the entire prediction chain for each new disease.
[0003] Therefore, there is an urgent need for a disease model construction method that can uniformly describe multiple disease processes, has configurable parameters, and supports platform-based expansion. Summary of the Invention
[0004] To solve or alleviate one or more of the above-mentioned technical problems, the present invention provides a method and system for constructing crop disease models based on the coupling of multiple environmental factors.
[0005] According to one aspect of the present invention, a method for constructing a crop disease model based on the coupling of multiple environmental factors is proposed, the method comprising: S1. Acquire environmental data, soil data, crop data, and pathogen data; S2. Based on the pathogen data, calculate the spore attachment success rate; the spore attachment success rate represents the probability that spores successfully land from the air onto the surface of crop leaves and remain there. S3. Divide the spore infection process into four stages: dormancy, germination, germ tube, and infection spike. Construct an infection state machine model. The infection state machine model is used to maintain discrete spore stage states and calculate continuous germination progress values and cumulative infection values based on the environmental data, crop data, and pathogen data, and then calculate the infection activity output. S4. Based on the soil data and the crop data, calculate multiple correction coefficients, including at least the water stress development correction coefficient, the nutrient stress development correction coefficient, the agricultural pesticide application pathogen mortality correction coefficient, and the carbon dioxide content stress correction coefficient. S5. Multiply the spore attachment success rate, the infection activity output, and the multiple correction coefficients to obtain the disease risk value.
[0006] Furthermore, the environmental data includes temperature, relative humidity, wind speed, precipitation, ultraviolet light intensity, and carbon dioxide content; the soil data includes soil volumetric water content; the crop data includes: crop variety resistance value, growth period information, leaf area index, leaf angle coefficient, wilting point water content, field capacity, drought stress lower limit threshold, drought resistance index, flood stress upper limit threshold, flood resistance index, actual crop uptake, and theoretical nutrient requirements; the pathogen data includes eight parameter tables: a wind speed attachment probability table, recording the probability of spores attaching to crop leaves at different wind speeds; a precipitation washout probability table, recording the probability of spores being washed away from leaves by rainwater at different rainfall intensities; and a germination humidity requirement table, recording the probability of spores being washed away from leaves by rainwater at different temperatures. The table lists the minimum relative humidity required for spores to remain active or complete infection at various stages under different temperatures, including dormancy threshold, germination threshold, germ tube stage threshold, and infection spike stage threshold; a germination probability table, recording the probability of spores successfully completing the germination stage at different temperatures; a germination duration table, recording the number of hours required for spores to complete the current stage at different temperatures, including the duration of each stage: dormancy, germination, germ tube, and infection spike stage; a dormancy-breaking time factor table, recording the time conversion factor required to break spore dormancy under different temperature and humidity conditions; a UV light mortality rate table, recording the probability of spore death per hour under different UV light intensities; and a pesticide pathogen mortality rate table, recording the lethal probability of pathogens for different pesticide varieties and concentrations.
[0007] Furthermore, the formula for calculating the spore attachment success rate A is as follows: A = Fv × (1 - Pr); Where Fv represents the wind speed attachment probability, which is obtained from the wind speed attachment probability table based on wind speed through a linear interpolation function; Pr represents the precipitation scour probability, which is obtained from the precipitation scour probability table based on precipitation amount through a linear interpolation function.
[0008] Furthermore, the calculation of infection activity output includes: The current stage of the spore is determined based on the infection state machine model; The germination progress value value11 and the cumulative infection value value21 are calculated based on the current stage of the spore. The germination progress value value11 represents the proportion of the spore's progress from the start of the current stage to the completion of the current stage. The cumulative infection value value21 represents the cumulative probability of successful infection from the start of spore germination to the current moment. The infection amount value31 after UV light correction is determined based on the cumulative infection value21 and the UV light intensity; the infection amount value31 after UV light correction represents the proportion of spores that still survive and can continue to complete infection after considering the killing effect of UV light on spores. The infection value after resistance correction (value31) was determined as the infection activity output based on the infection value after UV light correction and the crop variety resistance value.
[0009] Furthermore, the calculation of the germination progress value value11 and the cumulative infection value value21 based on the current stage of the spore includes: determining whether the spore is in an active, suspended, or dead state based on temperature, relative humidity, and drying time; when the spore is in an active state, calculating the current germination progress value value11 and the cumulative infection value value21 based on the current stage of the spore; when the spore is in a suspended state, the germination progress value value11 and the cumulative infection value value21 remain unchanged; when the spore is in a dead state, the germination progress value value11, the cumulative infection value value21, the infection amount after UV light correction value31, and the infection amount after resistance correction value41 are all reset to zero.
[0010] Furthermore, the formulas for calculating the germination progress value (value11) and the cumulative infection value (value21) are as follows: value11=clip(value11_old+dV1,0,1); value21=clip(value21_old+Pg×dV1,0,1); Where clip represents the clipping function, limiting the value to the range of 0 to 1; value11_old represents the germination progress value at the previous time step; value21_old represents the cumulative infection value at the previous time step; dV1 represents the germination progress increment, dV1 = 1 / (Hg) Dx), Hg represents the time required for germination, Dx represents the dormancy breaking time factor; Pg represents the germination probability.
[0011] Furthermore, the formulas for calculating the infection value31 after UV irradiation correction and the infection value41 after resistance correction are as follows: value31 = value21 × (1 M); value41 = Kres × value31; Where M represents the mortality rate from ultraviolet light exposure; Kres represents the crop resistance correction factor, Kres = (10-kx) / 10; and kx represents the crop variety resistance value.
[0012] Furthermore, the ultraviolet radiation mortality rate M is determined as follows: the ultraviolet radiation intensity at different depths within the canopy is calculated based on the Beer-Lambert law, and the ultraviolet radiation intensity of the i-th layer is the weighted average of the direct light intensity and the scattered light intensity of the i-th layer; the ultraviolet radiation mortality rate M of the i-th layer is obtained from the ultraviolet radiation mortality rate table based on the ultraviolet radiation intensity of the i-th layer through a linear interpolation function. The direct light intensity of the i-th layer is calculated as follows: Divide the canopy into W equal layers. For the i-th layer, the cumulative leaf area index L of the direct light intensity from the top of the canopy to that layer is... , Let I_direct represent the leaf area index. Then, the formula for calculating the direct light intensity I_direct of the i-th layer is: I_direct=PAR×exp( k×L); In the formula, PAR represents ultraviolet light intensity; exp represents an exponential function with the natural constant e as the base; and k represents the leaf angle coefficient. The intensity of scattered light from the i-th layer is calculated as follows: I_scatter = I_direct × S_i; In the formula, S_i represents the scattering ratio of the i-th layer, with S_i=0.1 at the top of the canopy, S_i=0.9 at the bottom of the canopy, and S_i in the middle of the canopy is calculated by linear interpolation.
[0013] Furthermore, the water stress development correction factor Cw is calculated as follows: 1) When the soil volumetric water content Sm is less than the lower limit threshold of drought stress Sdrought, the calculation formula is: Cw = max(1-(Sdrought-Sm) / (Sfc-Swp) (1 / Ddrought), 0); 2) When the soil volumetric water content Sm is greater than the upper limit threshold of flood stress Sflood, the calculation formula is: Cw = max(1-(Sm-Sflood) / (Sfc-Swp) (1 / Dflood),0); 3) When Sdrought ≤ Sm ≤ Sflood, Cw = 1; The formula for calculating the nutritional stress developmental correction factor Cn is: Cn = max(1-|UD| / D,0); The formula for calculating the mortality correction factor Ce for pathogens in agricultural pesticide application is as follows: Ce = 1 - Mnatural; The formula for calculating the carbon dioxide stress correction factor Cp is: Cp = 1 + α ln(Patm / 420); In the formula, Sfc represents field water holding capacity; Swp represents wilting point water content; Ddrought represents drought resistance index; Dflood represents flood resistance index; U represents actual crop absorption; D represents theoretical nutrient requirement; Mnatural represents the mortality rate of pathogens from agricultural pesticide application; α represents the carbon dioxide sensitivity of pathogens; and Patm represents carbon dioxide content.
[0014] According to one aspect of the present invention, a crop disease model construction system based on multi-environmental factor coupling is proposed, the system being used to implement the above-described crop disease model construction method based on multi-environmental factor coupling; the system includes: The data acquisition module is used to acquire environmental data, soil data, crop data, and pathogen data. The disease risk construction module includes a data calculation submodule and a risk calculation submodule. The data calculation submodule calculates the spore attachment success rate based on the pathogen data. The spore attachment success rate represents the probability that spores successfully land on the surface of crop leaves from the air and remain there. The spore infection process is divided into four stages: dormancy, germination, germ tube, and infection nail stage. An infection state machine model is constructed to maintain the spore stage state and calculate continuous germination progress and cumulative infection values based on the environmental data, crop data, and pathogen data, thereby calculating the infection activity output. Based on the soil data and crop data, multiple correction coefficients are calculated, including at least a water stress development correction coefficient, a nutrient stress development correction coefficient, an agricultural pesticide pathogen mortality correction coefficient, and a carbon dioxide content stress correction coefficient. The risk calculation submodule multiplies the spore attachment success rate, the infection activity output, and the multiple correction coefficients to obtain the disease risk value.
[0015] The embodiments of the present invention have the following technical effects: 1) Unified framework, reuse of multiple diseases: By abstracting the general processes such as spore attachment, infection state machine, and multi-factor correction, any disease can reuse the same model framework by configuring the corresponding pathogen parameter table, and the development cycle of new diseases is shortened from 3 months to 2 weeks. 2) Parametric configuration and high generalization ability: The model behavior is driven by linear interpolation, which can quickly adapt to different climate zones and different crop varieties; 3) Supports platform-based deployment: The unified input and output interface enables the model to be embedded in agricultural SaaS systems and government-level disaster early warning systems; 4) Combining mechanism and experience: The state machine model reflects the biological process of pathogens, and the pathogen parameter table incorporates field experience data, which is both interpretable and adaptable. Attached Figure Description
[0016] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0017] Figure 1 This is a flowchart of a method for constructing a crop disease model based on the coupling of multiple environmental factors, as described in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of a crop disease model construction system based on the coupling of multiple environmental factors, as described in an embodiment of the present invention. Detailed Implementation
[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are part of this invention.
[0019] This invention proposes a method for constructing a crop disease model based on the coupling of multiple environmental factors, such as... Figure 1 As shown, the method includes: S1. Acquire environmental data, soil data, crop data, and pathogen data; S2. Based on the pathogen data, calculate the spore attachment success rate; the spore attachment success rate represents the probability that spores successfully land from the air onto the surface of crop leaves and remain there. S3. Divide the spore infection process into four stages: dormancy, germination, germ tube, and infection spike. Construct an infection state machine model. The infection state machine model is used to maintain discrete spore stage states and calculate continuous germination progress values and cumulative infection values based on the environmental data, crop data, and pathogen data, and then calculate the infection activity output. S4. Based on the soil data and the crop data, calculate multiple correction coefficients, including at least the water stress development correction coefficient, the nutrient stress development correction coefficient, the agricultural pesticide application pathogen mortality correction coefficient, and the carbon dioxide content stress correction coefficient. S5. Multiply the spore attachment success rate, the infection activity output, and the multiple correction coefficients to obtain the disease risk value.
[0020] The method begins with S1, in which environmental data, soil data, crop data, and pathogen data are acquired.
[0021] According to embodiments of the present invention, environmental data, including temperature, relative humidity, wind speed, precipitation, ultraviolet radiation intensity, and carbon dioxide content, are acquired from field weather stations on an hourly basis. Temperature is expressed in degrees Celsius, relative humidity as a percentage, wind speed in meters per second, precipitation in millimeters, and ultraviolet radiation intensity in watts per square meter. Carbon dioxide content varies across different regions, for example, between eastern and western areas. Furthermore, many facility-based agricultural projects release carbon dioxide to achieve high yields, thus requiring real-time monitoring of the actual carbon dioxide content.
[0022] Soil data is obtained from soil sensors, and the soil data includes soil volumetric water content.
[0023] Crop data are derived from variety databases, crop parameter libraries, soil nutrient testing, crop growth models, and agricultural records, and include the following 14 parameters: crop variety resistance value kx; growth period information, including tillering stage, jointing stage, booting stage, grain-filling stage, and maturity stage; leaf area index (LAI); leaf angle coefficient (k); wilting point water content (Swp); field capacity (Sfc); lower limit threshold for drought stress (Sdrought); drought resistance index (Ddrought); upper limit threshold for flood stress (Sflood); flood resistance index (Dflood); actual crop uptake (U), in kilograms per acre; and theoretical nutrient requirement (D), in kilograms per acre. Among them, the variety database contains variety specifications, from which crop variety resistance values kx and leaf angle coefficient k can be determined; the crop parameter library obtained through experimental summary can provide wilting point water content Swp, field water holding capacity Sfc, drought stress lower limit threshold Sdrought, drought resistance index Ddrought, flood stress upper limit threshold Sflood, and flood resistance index Dflood; soil nutrient detection obtained from experimental analysis can provide the actual crop uptake U; the crop growth model obtained through mathematical model simulation based on meteorological data, soil data, and crop parameters can provide growth period information and theoretical nutrient requirements D; and agricultural records obtained from the agricultural production process can provide leaf area index LAI.
[0024] The pathogen data consists of a complete set of parameter tables pre-stored in the database for each disease, comprising the following eight parameter tables: 1) Wind speed attachment probability table, recording the probability of spores attaching to crop leaves at different wind speeds; 2) Rainfall washout probability table, recording the probability of spores being washed away from leaves by rainwater at different rainfall intensities; 3) Germination humidity requirement table, recording the minimum relative humidity required for spores to remain active or complete infection at different stages under different temperatures, including dormancy threshold, germination threshold, germ tube stage threshold, and infection spike stage threshold; 4) Germination probability table, recording the probability of spores successfully completing the germination stage at different temperatures; 5) Germination duration table, recording the number of hours required for spores to complete the current stage at different temperatures, including the duration of each stage: dormancy, germination, germ tube, and infection spike stage; 6) Dormancy breaking time factor table, recording the time conversion factor required to break spore dormancy under different temperature and humidity conditions; 7) Ultraviolet light mortality rate table, recording the probability of spore death per hour under different ultraviolet light intensities; 8) Agricultural pesticide pathogen mortality rate table, recording the lethal probability of different pesticide varieties and concentrations for pathogens.
[0025] Then, in S2, based on the pathogen data, the spore attachment success rate is calculated; the spore attachment success rate represents the probability that spores successfully land from the air onto the surface of crop leaves and are able to remain there.
[0026] According to embodiments of the present invention, the first step in disease occurrence is that pathogen spores must reach the crop leaves. However, spores do not fly there on their own; they are spread by wind. Wind can both help spores attach to the leaves and potentially blow them away. More importantly, rainwater can both cause spores to settle and wash away spores already on the leaves. The spore attachment success rate is calculated by comprehensively considering the promoting effect of wind and the washing effect of rain, determining how many spores ultimately remain on the leaves.
[0027] The formula for calculating the spore attachment success rate A is: A = Fv × (1 - Pr); where Fv represents the wind speed attachment probability, that is, the probability that the wind blows the spores onto the leaf and they are intercepted by the leaf; Pr represents the precipitation washout probability, that is, the probability that rainwater falls on the leaf and washes away the existing spores.
[0028] The wind speed attachment probability Fv is obtained from the wind speed attachment probability table using a linear interpolation function based on the wind speed. When the actual wind speed is between two recorded points, the linear interpolation formula is used as follows: Suppose there are two adjacent recorded points in the wind speed attachment probability table, the wind speed of the left recorded point is ws_L, and the corresponding attachment probability is Fv_L; the wind speed of the right recorded point is ws_R, and the corresponding attachment probability is Fv_R; the actual wind speed is ws, and ws_L≤ws≤ws_R, then the linear interpolation formula for the wind speed attachment probability Fv corresponding to the actual wind speed is: Fv=Fv_L+(Fv_R Fv_L)× ; The precipitation erosion probability Pr is obtained from the precipitation erosion probability table based on precipitation amount using the same linear interpolation method.
[0029] Then, in S3, the spore infection process is divided into four stages: dormancy, germination, germ tube, and infection spike. An infection state machine model is constructed. The infection state machine model is used to maintain the spore stage state and calculate continuous germination progress values and cumulative infection values based on the environmental data, crop data, and pathogen data, and then calculate the infection activity output. This includes: determining the current stage of the spore based on an infection state machine model; calculating a germination progress value (value11) and a cumulative infection value (value21) based on the current stage of the spore; the germination progress value (value11) represents the proportion of the spore's progress from the start of the current stage to the completion of the current stage; the cumulative infection value (value21) represents the cumulative probability of successful infection from the start of spore germination to the current moment; determining the infection amount (value31) after UV light correction based on the cumulative infection value (value21) and UV light intensity; the infection amount (value31) after UV light correction represents the proportion of spores that survive and can continue to complete infection after considering the killing effect of UV light on spores; and determining the infection amount (value41) after resistance correction based on the infection amount (value31) after UV light correction and the crop variety resistance value as the output of infection activity.
[0030] According to an embodiment of the present invention, the calculation of the infection activity output I adopts a state machine model to simulate the complete biological process of spores from dormancy to germination, formation of germ tubes, formation of infection nails, and completion of infection. The system maintains state variables for each spatiotemporal grid point, including the germination progress parameter value11, the cumulative infection parameter value21, the infection amount after UV light correction value31, the infection amount after resistance correction value41, and the drying time drytime. Among them, the drying time drytime is used to record the duration of spores in a dry environment and is a key parameter for determining whether spores are dead.
[0031] Specifically, the current stage of the spore is first determined based on the infection state machine model; and the spore is judged to be in an active, suspended, or dead state based on temperature, relative humidity, and drying time; when the spore is in an active state, the current germination progress value value11 and the cumulative infection value value21 are calculated based on the current stage of the spore; when the spore is in a suspended state, the germination progress value value11 and the cumulative infection value value21 remain unchanged; when the spore is in a dead state, the germination progress value value11, the cumulative infection value value21, the infection amount after UV light correction value31, and the infection amount after resistance correction value41 are all reset to zero.
[0032] For dormant spores, they are in an active state when the relative humidity is greater than or equal to 80% (80% is the dormant threshold obtained from the germination humidity requirement table), and the drying time is reset to zero; when the relative humidity is less than 80% (which is also the drying death trigger threshold), the drying time begins to accumulate.
[0033] For germinating spores, they are in an active state when the relative humidity is greater than or equal to 95% (95% is the germination threshold obtained from the germination humidity requirement table); they enter a paused state when the relative humidity is between 80% and 95%, and the drying time does not accumulate; when the relative humidity is lower than 80% (80% is also the drying death trigger threshold), the drying time begins to accumulate.
[0034] For germ tube spores, they are in an active state when the relative humidity is greater than or equal to 90% (90% is the threshold for germ tube spores obtained from the germination humidity requirement table); when the relative humidity is between 80% and 90%, they enter a paused state and the drying time does not accumulate; when the relative humidity is below 80% (80% is also the threshold for triggering drying death), the drying time begins to accumulate.
[0035] For spores in the nail stage, infection is completed when the relative humidity is greater than or equal to 95% (95% is the infection period threshold obtained from the germination humidity requirement table); when the relative humidity is between 80% and 95%, it enters a pause state and the drying time does not accumulate; when the relative humidity is below 80% (80% is also the drying death trigger threshold), the drying time begins to accumulate.
[0036] The spore state assessment function determines whether a spore is dead based on its current stage and the accumulated drying time (drytime). Dormant spores are considered dead after 24 hours of continuous drying; germinating spores after 3 hours; germ tube spores after 2 hours; and infection spike spores after 1 hour. Once death is determined, the system resets all infection state variables (value11, value21, value31, and value41) to zero, and the infection process terminates.
[0037] The germination progress parameter `value11` is a time progress variable representing the percentage of spore growth from the start of the current stage to the completion of that stage, with a value ranging from 0 to 1. The calculation formula is: value11=clip(value11_old+dV1,0,1); Where clip represents the clipping function, limiting the value to the range of 0 to 1; value11_old represents the germination progress value at the previous time step; dV1 = 1 / (Hg Dx represents the germination progress increment, Hg is the germination time required, which is obtained from the germination time table based on temperature using a linear interpolation function; Dx is the dormancy breaking time factor, which is obtained from the dormancy breaking time factor table using a linear interpolation function. The higher the humidity, the smaller the Dx value and the faster the germination, and the lower the humidity, the larger the Dx value and the slower the germination.
[0038] The cumulative infection parameter value21 is the result of multiplying value11 by the germination probability Pg. The calculation formula is: value21=clip(value21_old+Pg×dV1,0,1); Where value21_old represents the cumulative infection value at the previous time step; the germination probability Pg is obtained from the germination probability table based on temperature using a linear interpolation function. The difference between value21 and value11 is that value11 only cares about how much time has progressed, not whether this progress is effective; value21 cares about how much of this time progress is converted into the actual probability of successful infection under temperature regulation.
[0039] The UV-corrected infection value (value31) is the result of multiplying value21 by the UV-induced survival rate. The calculation formula is: value31 = value21 × (1 M); Where M represents the mortality rate from ultraviolet light. The physical meaning of the infection rate value31 after ultraviolet light correction is: among the spores that have completed the infection process, some will be killed by ultraviolet light, and only the surviving ones can continue.
[0040] The ultraviolet (UV) radiation mortality rate M is determined as follows: The UV radiation intensity at different depths within the canopy is calculated based on the Beer-Lambert law. The UV radiation intensity of the i-th layer is a weighted average of the direct light intensity and the scattered light intensity of the i-th layer. Based on the UV radiation intensity of the i-th layer, the UV radiation mortality rate M of that layer is obtained from the UV radiation mortality rate table through a linear interpolation function.
[0041] Specifically, the incident ultraviolet light intensity value PAR at the top of the canopy is first obtained from environmental data; then the leaf area index (LAI) is determined according to the crop growth stage, for example: LAI is 1.5 for the tillering stage, 3 for the jointing stage, 4 for the booting stage, 5.5 for the grain-filling stage, and 3.5 for the maturity stage; the leaf angle coefficient k is set according to the crop type, usually 0.6 for rice and other gramineous crops, and usually 0.8 for broadleaf crops.
[0042] Divide the canopy into W equal layers, each representing 1 / W of the canopy height. For example, W=10. For the i-th layer, the cumulative leaf area index L from the top of the canopy to that layer is... The formula for calculating the direct light intensity I_direct of the i-th layer is: I_direct=PAR×exp( k×L); In the formula, exp represents an exponential function with the natural constant e as its base.
[0043] The calculation of scattered light intensity is based on the following scattering ratio model: the scattering ratio S_i increases linearly with the depth of the canopy. At the top of the canopy, i.e., when the depth is zero, the scattering ratio S_i is 0.1, and at the bottom of the canopy, i.e., when the depth is 1, the scattering ratio S_i is 0.9. For intermediate depths, S_i is calculated by linear interpolation; then the scattered light intensity of the i-th layer is I_scatter = I_direct × S_i.
[0044] The infection value (value41) after resistance correction is the result of multiplying value31 by the crop resistance correction factor Kres. The calculation formula is as follows: value41 = Kres × value31; Where Kres = (10-kx) / 10; kx represents the crop variety resistance value, used to quantify the crop variety's ability to resist infection by a certain pathogen, and its value is usually in the range of 0 to 10. The larger the kx value, the stronger the resistance; the smaller the value, the more susceptible the disease. The physical meaning of the infection value value41 after resistance correction is: even if the spores are not killed by ultraviolet light, the crop's own immune system can still prevent some infection. Finally, value41 is entered into the main model as the infection activity output I.
[0045] Then, in S4, based on the soil data and the crop data, a number of correction coefficients are calculated, including at least a water stress development correction coefficient, a nutrient stress development correction coefficient, an agricultural pesticide pathogen mortality correction coefficient, and a carbon dioxide content stress correction coefficient.
[0046] According to an embodiment of the present invention, the water stress development correction coefficient Cw is calculated based on the relationship between soil volumetric water content and the crop's water stress tolerance range. The current soil volumetric water content Sm is obtained from a soil sensor, and the wilting point water content Swp, field capacity Sfc, drought stress lower limit threshold Sdrought, flood stress upper limit threshold Sflood, drought resistance index Ddrought, and flood resistance index Dflood are obtained from a crop parameter database.
[0047] When the soil volumetric water content Sm is less than the drought stress lower limit threshold Sdrought, the crop is under drought stress. The formula for calculating the water stress development correction coefficient is as follows: Cw = max(1-(Sdrought-Sm) / (Sfc-Swp) (1 / Ddrought), 0); When the soil volumetric water content Sm is greater than the upper limit threshold of flood stress Sflood, the crop is under flood stress. The formula for calculating the water stress development correction coefficient is: Cw = max(1-(Sm-Sflood) / (Sfc-Swp) (1 / Dflood),0); In other cases, when Sdrought≤Sm≤Sflood, the crop moisture condition is suitable, and Cw = 1.
[0048] Furthermore, for situations involving persistent drought or flooding, this embodiment adds an hourly cumulative mechanism. When the soil volumetric water content Sm is less than the lower limit threshold of drought stress Sdrought, the crop is under drought stress; when the soil volumetric water content Sm is greater than the upper limit threshold of flood stress Sflood, the crop is under flood stress. Starting from the first hour of entering the stress state, the current water stress development correction coefficient is multiplied by a decay factor slightly less than 1; when the water conditions return to normal, it is gradually multiplied by a recovery factor slightly greater than 1 hourly until it recovers to 1.
[0049] The nutrient stress development correction coefficient Cn is calculated based on the balance between crop nutrient uptake and demand. Actual crop uptake (U) is obtained from soil nutrient analysis, and the theoretical nutrient demand (D) for that growth stage is obtained from a crop growth model. The formula for calculating the nutrient stress development correction coefficient is: Cn = max(1-|UD| / D,0); Where |U - D| represents the absolute value of the difference between U and D. When the amount absorbed is exactly equal to the amount required, i.e., U=D, Cn=1, indicating that the crop has the best nutritional status and the strongest disease resistance; when |U - D|≥D, Cn=0, indicating that the amount absorbed is seriously insufficient or excessive.
[0050] Furthermore, the nutritional stress developmental correction coefficient Cn can be calculated using a product combination of multiple nutrient factors. When considering nitrogen, phosphorus, and potassium simultaneously, the corresponding nutritional stress developmental correction coefficients Cn_N, Cn_P, and Cn_K are calculated separately. The comprehensive nutritional stress developmental correction coefficient is then: Cn = Cn_N × Cn_P × Cn_K, or the minimum value of the three can be taken: Cn = min(Cn_N, Cn_P, Cn_K).
[0051] The formula for calculating the mortality correction coefficient Ce for pesticide-treated pathogens is: Ce = 1 - Mnatural; where Mnatural is the mortality rate of pesticide-treated pathogens, ranging from 0 to 1. It comes from the pesticide-treated pathogen mortality rate table and is a static parameter describing the characteristics of pathogens. The pesticide-treated pathogen mortality rate table can be configured based on field natural enemy population monitoring data or historical statistical data.
[0052] The formula for calculating the carbon dioxide stress correction factor Cp is: Cp = 1 + α ln(Patm / 420); where α represents the carbon dioxide sensitivity coefficient of the pathogen, and its value range is configured according to the type of pathogen. α>0 indicates high concentration promotes growth, and α<0 indicates high concentration inhibits growth. Patm represents the carbon dioxide content in the air, in ppm; 420 is the standard carbon dioxide content constant, in ppm. The design basis of the carbon dioxide stress correction coefficient Cp is that pathogens have different sensitivities to carbon dioxide content. Generally speaking, higher carbon dioxide content is conducive to spore germination and infection, while lower carbon dioxide content is unfavorable to pathogen activity.
[0053] Then, in S5, the spore attachment success rate, infection activity output, and multiple correction coefficients are multiplied together to obtain the disease risk value.
[0054] According to an embodiment of the present invention, the formula for calculating the disease risk value is Risk: Risk = A × I × Cw × Cn × Ce × Cp.
[0055] The disease risk value for multiple layers can be calculated according to the above formula based on the canopy location. For example, it can correspond to the root position, one-third height position, sub-canopy position, and top canopy position respectively.
[0056] The calculated disease risk values are then stored in a database for subsequent risk map creation, early warning information dissemination, or prevention and control decision support.
[0057] This invention can be deployed in agricultural meteorological service stations, smart agriculture cloud platforms, and drone-based plant protection decision-making systems. Through a standard API interface, it can simultaneously run multiple disease models, including rice blast, wheat scab, corn rust, and potato late blight, outputting hourly or daily risk warnings to guide precise pesticide application.
[0058] This invention also proposes a crop disease model construction system based on the coupling of multiple environmental factors. This system is used to implement the crop disease model construction method based on the coupling of multiple environmental factors described in the above embodiments. Figure 2 As shown, the system includes: The data acquisition module 210 is used to acquire environmental data, soil data, crop data, and pathogen data; The disease risk construction module 220 includes a data calculation submodule 2210 and a risk calculation submodule 2220. The data calculation submodule 2210 calculates the spore attachment success rate based on the pathogen data. The spore attachment success rate represents the probability that spores successfully land on the surface of crop leaves from the air and remain there. The spore infection process is divided into four stages: dormancy, germination, germ tube, and infection nail stage. An infection state machine model is constructed, which maintains the spore stage state and calculates continuous germination progress and cumulative infection values based on the environmental data, crop data, and pathogen data, thereby calculating the infection activity output. Based on the soil data and crop data, multiple correction coefficients are calculated, including at least a water stress development correction coefficient, a nutrient stress development correction coefficient, an agricultural pesticide pathogen mortality correction coefficient, and a carbon dioxide content stress correction coefficient. The risk calculation submodule 2220 multiplies the spore attachment success rate, the infection activity output, and the multiple correction coefficients to obtain the disease risk value.
[0059] It should be noted that the function of the crop disease model construction system based on the coupling of multiple environmental factors described in the embodiments of the present invention can be explained by the aforementioned crop disease model construction method based on the coupling of multiple environmental factors. Therefore, for the parts not described in detail in the system embodiments, please refer to the above method embodiments, and they will not be repeated here.
[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.
Claims
1. A method for constructing a crop disease model based on the coupling of multiple environmental factors, characterized in that, include: S1. Acquire environmental data, soil data, crop data, and pathogen data; S2. Based on the pathogen data, calculate the spore attachment success rate; the spore attachment success rate represents the probability that spores successfully land from the air onto the surface of crop leaves and remain there. S3. Divide the spore infection process into four stages: dormancy, germination, germ tube, and infection spike. Construct an infection state machine model. The infection state machine model is used to maintain the spore stage state and calculate continuous germination progress values and cumulative infection values based on the environmental data, crop data, and pathogen data, and then calculate the infection activity output. S4. Based on the soil data and the crop data, calculate multiple correction coefficients, including at least the water stress development correction coefficient, the nutrient stress development correction coefficient, the agricultural pesticide application pathogen mortality correction coefficient, and the carbon dioxide content stress correction coefficient. S5. Multiply the spore attachment success rate, the infection activity output, and the multiple correction coefficients to obtain the disease risk value.
2. The method for constructing a crop disease model based on the coupling of multiple environmental factors according to claim 1, characterized in that, The environmental data includes temperature, relative humidity, wind speed, precipitation, ultraviolet light intensity, and carbon dioxide content; the soil data includes soil volumetric water content; the crop data includes: crop variety resistance value, growth period information, leaf area index, leaf angle coefficient, wilting point water content, field capacity, lower threshold of drought stress, drought resistance index, upper threshold of flood stress, flood resistance index, actual crop uptake, and theoretical nutrient requirements; the pathogen data includes eight parameter tables: a wind speed attachment probability table, recording the probability of spores attaching to crop leaves at different wind speeds; a precipitation washout probability table, recording the probability of spores being washed away from leaves by rainwater at different rainfall intensities; and a germination humidity requirement table, recording the probability of spores being washed away from leaves at different temperatures. The table lists the minimum relative humidity required for spores to remain active or complete infection at each stage, including the dormancy threshold, germination threshold, germ tube threshold, and infection spike threshold; a germination probability table, recording the probability of spores successfully completing the germination stage at different temperatures; a germination duration table, recording the number of hours required for spores to complete the current stage at different temperatures, including the duration of each stage: dormancy, germination, germ tube, and infection spike; a dormancy-breaking time factor table, recording the time conversion factor required to break spore dormancy under different temperature and humidity conditions; a UV light mortality rate table, recording the probability of spore death per hour under different UV light intensities; and a pesticide pathogen mortality rate table, recording the lethal probability of different pesticide varieties and concentrations for pathogens.
3. The method for constructing a crop disease model based on the coupling of multiple environmental factors according to claim 2, characterized in that, The formula for calculating the spore attachment success rate A is as follows: A = Fv × (1 - Pr); Where Fv represents the wind speed attachment probability, which is obtained from the wind speed attachment probability table based on wind speed through a linear interpolation function; Pr represents the precipitation scour probability, which is obtained from the precipitation scour probability table based on precipitation amount through a linear interpolation function.
4. The method for constructing a crop disease model based on the coupling of multiple environmental factors according to claim 3, characterized in that, The calculated infection activity output includes: The current stage of the spore is determined based on the infection state machine model; The germination progress value value11 and the cumulative infection value value21 are calculated based on the current stage of the spore. The germination progress value value11 represents the proportion of the spore's progress from the start of the current stage to the completion of the current stage. The cumulative infection value value21 represents the cumulative probability of successful infection from the start of spore germination to the current moment. The infection amount value31 after UV light correction is determined based on the cumulative infection value21 and the UV light intensity; the infection amount value31 after UV light correction represents the proportion of spores that still survive and can continue to complete infection after considering the killing effect of UV light on spores. The infection value after resistance correction (value31) was determined as the infection activity output based on the infection value after UV light correction and the crop variety resistance value.
5. The method for constructing a crop disease model based on the coupling of multiple environmental factors according to claim 4, characterized in that, The calculation of germination progress value (value11) and cumulative infection value (value21) based on the current stage of the spore includes: determining whether the spore is active, suspended, or dead based on temperature, relative humidity, and drying time; when the spore is active, calculating the current germination progress value (value11) and cumulative infection value (value21) based on the current stage of the spore; when the spore is suspended, the germination progress value (value11) and cumulative infection value (value21) remain unchanged; when the spore is dead, the germination progress value (value11), cumulative infection value (value21), infection amount after UV correction (value31), and infection amount after resistance correction (value41) are all reset to zero.
6. The method for constructing a crop disease model based on the coupling of multiple environmental factors according to claim 5, characterized in that, The formulas for calculating the germination progress value (value11) and the cumulative infection value (value21) are as follows: value11=clip(value11_old+dV1,0,1); value21=clip(value21_old+Pg×dV1,0,1); Where clip represents the clipping function, limiting the value to the range of 0 to 1; value11_old represents the germination progress value at the previous time step; value21_old represents the cumulative infection value at the previous time step; dV1 represents the germination progress increment, dV1 = 1 / (Hg) Dx), Hg represents the time required for germination, Dx represents the dormancy breaking time factor; Pg represents the germination probability.
7. The method for constructing a crop disease model based on the coupling of multiple environmental factors according to claim 6, characterized in that, The formulas for calculating the infection value (value31) after UV irradiation correction and the infection value (value41) after resistance correction are as follows: value31=value21×(1 M); value41 = Kres × value31; Where M represents the mortality rate from ultraviolet light exposure; Kres represents the crop resistance correction factor, Kres = (10-kx) / 10; and kx represents the crop variety resistance value.
8. The method for constructing a crop disease model based on the coupling of multiple environmental factors according to claim 7, characterized in that, The ultraviolet radiation mortality rate M is determined as follows: the ultraviolet radiation intensity at different depths within the canopy is calculated based on the Beer-Lambert law, and the ultraviolet radiation intensity of the i-th layer is the weighted average of the direct light intensity and the scattered light intensity of the i-th layer; the ultraviolet radiation mortality rate M of the i-th layer is obtained from the ultraviolet radiation mortality rate table based on the ultraviolet radiation intensity of the i-th layer through a linear interpolation function. The direct light intensity of the i-th layer is calculated as follows: Divide the canopy into W equal layers. For the i-th layer, the cumulative leaf area index L of the direct light intensity from the top of the canopy to that layer is... , Let I_direct represent the leaf area index. Then, the formula for calculating the direct light intensity I_direct of the i-th layer is: I_direct=PAR×exp( k×L); In the formula, PAR represents ultraviolet light intensity; exp represents an exponential function with the natural constant e as the base; and k represents the leaf angle coefficient. The intensity of scattered light from the i-th layer is calculated as follows: I_scatter = I_direct × S_i; In the formula, S_i represents the scattering ratio of the i-th layer, with S_i=0.1 at the top of the canopy, S_i=0.9 at the bottom of the canopy, and S_i in the middle of the canopy is calculated by linear interpolation.
9. The method for constructing a crop disease model based on the coupling of multiple environmental factors according to claim 8, characterized in that, The water stress development correction factor Cw is calculated as follows: 1) When the soil volumetric water content Sm is less than the lower limit threshold of drought stress Sdrought, the calculation formula is: Cw = max(1-(Sdrought-Sm) / (Sfc-Swp) (1 / Ddrought), 0); 2) When the soil volumetric water content Sm is greater than the upper limit threshold of flood stress Sflood, the calculation formula is: Cw = max(1-(Sm-Sflood) / (Sfc-Swp) (1 / Dflood),0); 3) When Sdrought ≤ Sm ≤ Sflood, Cw = 1; The formula for calculating the nutritional stress developmental correction factor Cn is: Cn = max(1-|UD| / D,0); The formula for calculating the mortality correction factor Ce for pathogens in agricultural pesticide application is as follows: Ce = 1 - Mnatural; The formula for calculating the carbon dioxide stress correction factor Cp is: Cp=1+ α ln(Patm / 420); In the formula, Sfc represents field water holding capacity; Swp represents wilting point water content; Ddrought represents drought resistance index; Dflood represents flood resistance index; U represents actual crop absorption; D represents theoretical nutrient requirements; Mnatural represents the mortality rate of pathogens from agricultural pesticide application; α represents the carbon dioxide sensitivity coefficient of pathogens; and Patm represents carbon dioxide content.
10. A crop disease model construction system based on the coupling of multiple environmental factors, characterized in that, The system is used to implement the crop disease model construction method based on the coupling of multiple environmental factors as described in any one of claims 1-9; the system includes: The data acquisition module is used to acquire environmental data, soil data, crop data, and pathogen data; The disease risk construction module includes a data calculation submodule and a risk calculation submodule. The data calculation submodule calculates the spore attachment success rate based on the pathogen data. The spore attachment success rate represents the probability that spores successfully land on the surface of crop leaves from the air and remain there. The spore infection process is divided into four stages: dormancy, germination, germ tube, and infection nail stage. An infection state machine model is constructed to maintain the spore stage state and calculate continuous germination progress and cumulative infection values based on the environmental data, crop data, and pathogen data, thereby calculating the infection activity output. Based on the soil data and crop data, multiple correction coefficients are calculated, including at least a water stress development correction coefficient, a nutrient stress development correction coefficient, an agricultural pesticide pathogen mortality correction coefficient, and a carbon dioxide content stress correction coefficient. The risk calculation submodule multiplies the spore attachment success rate, the infection activity output, and the multiple correction coefficients to obtain the disease risk value.