Wheat stripe rust prediction method
By using the BP neural network model to predict the overall incidence trend of wheat stripe rust during the wheat rebirth and determining the meteorological suitability of the disease during the high incidence of the disease, the combination of long-term and short-term forecasts of wheat stripe rust is achieved, and the problems of compact forecasting time and failure to achieve long-term and short-term forecasts in the existing technology are solved, and the accuracy and timeliness of forecasting are improved.
Patent Information
- Application Number
- CN202510198748.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-21
- Publication Date
- 2025-05-09
AI Technical Summary
The existing methods for predicting wheat stripe rust are mainly concentrated in the wheat flowering period, with compact forecasting time and occurrence period, insufficient time for reserved disease prevention and control, and failure to achieve a combination of long-term and short-term forecasts.
The target meteorological factor was obtained during the wheat greening period and the iteratively trained BP neural network model was used to predict the overall incidence trend of wheat stripe rust. During the high incidence of wheat diseases, we obtain meteorological elements in the future period and determine the meteorological suitability of disease, so as to predict the probability of wheat stripe rust in the future period.
Long-term trend forecast and short-term dynamic forecast of wheat stripe rust have been achieved, combined with long-term and short-term forecasts, improving the accuracy and timeliness of forecasts, providing sufficient preparation time for wheat disease prevention and control.
Smart Images
Figure CN119963005A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of plant disease prediction, and in particular to a method for predicting wheat stripe rust. Background Art
[0002] Wheat stripe rust, commonly known as "jaundice", is a typical climate-type disease. Its pathogen spores can be spread over long distances with high-altitude air currents. Once an outbreak occurs, it can quickly spread over a large area. It has the characteristics of high occurrence intensity and wide epidemic range. It is one of the most important diseases in wheat production.
[0003] The factors affecting the spread of wheat stripe rust are very complex. The timing and extent of its occurrence depend on the wheat variety, stripe rust fungus, environmental conditions and their interactions. When both the pathogen and the host have the potential for epidemic, environmental conditions, especially meteorological conditions, become the dominant factor in the epidemic. That is, climate factors are an important factor in determining the speed and scope of the spread of wheat stripe rust. Wheat is prone to stripe rust under warm, rainy, dewy and humid conditions from the greening period to the flowering period.
[0004] In the past, the forecast of stripe rust mostly used the multivariate linear regression equation method, which was a single method. In recent years, with the development of machine learning, new technical methods such as random forest and decision tree models have begun to be applied to the forecast of wheat stripe rust. However, these methods did not make forecasts in stages, and most of them only made short-term forecasts during the wheat flowering period. The forecast time and occurrence period were tight, and there was insufficient time reserved for disease prevention and control.
[0005] Studies have shown that future climate change has a clear warming trend, which is beneficial for the safe wintering of stripe rust and the shortening of its incubation period. The frequency of stripe rust pandemics and severe damage is likely to increase significantly. Therefore, how to design a prediction method for wheat stripe rust to achieve a combined long-term and short-term forecast is crucial to the growth of wheat under future climate. Summary of the invention
[0006] The purpose of this application is to provide a wheat stripe rust prediction method that can achieve a combined forecast of long-term and short-term forecasts.
[0007] To achieve the above objectives, this application provides the following solutions:
[0008] The present application provides a method for predicting wheat stripe rust, comprising:
[0009] During the wheat greening and rooting period, target meteorological factors are obtained; the target meteorological factors include the meteorological factors of the current year and the meteorological factors of the previous year; the wheat greening and rooting period is March;
[0010] Inputting the target meteorological factors into a stripe rust prediction model to predict the overall occurrence trend of wheat stripe rust in that year; the stripe rust prediction model is obtained by iteratively training a BP neural network;
[0011] During the high incidence period of wheat diseases, obtain the meteorological elements of each day in the future period of the year; the meteorological elements include: temperature, precipitation, relative humidity and sunshine hours; the future period is within 3 days from the first day of the high incidence period of wheat diseases; the high incidence period of wheat diseases is April;
[0012] Determine the meteorological suitability of the disease in the future period of the year according to the meteorological elements of each day in the future period of the year; the meteorological suitability of the disease is used to indicate whether the meteorological conditions in the future period are suitable for the spread and development of wheat stripe rust;
[0013] The prediction result is obtained based on the overall occurrence trend of wheat stripe rust in the current year and the meteorological suitability of the disease; the prediction result includes the probability of occurrence of wheat stripe rust in the future period of the current year.
[0014] Optionally, the method for determining the stripe rust prediction model specifically includes:
[0015] Acquire training data; the training data includes: meteorological factors in a historical period and the corresponding overall occurrence trend of wheat stripe rust;
[0016] Dividing the training data to obtain training learning samples and forecast samples;
[0017] Inputting the training learning samples into the BP neural network for iterative training, and obtaining a trained BP neural network model when a set number of iterations is reached;
[0018] The trained BP neural network model is verified according to the forecast sample to obtain the forecast accuracy; the forecast accuracy is obtained by comparing the overall occurrence trend of wheat stripe rust in the forecast sample with the overall occurrence trend of wheat stripe rust output by the trained BP neural network model;
[0019] If the prediction accuracy reaches the set accuracy, the trained BP neural network model is determined as the stripe rust prediction model;
[0020] If the prediction accuracy does not reach the set accuracy, the number of the training learning samples is increased, and the BP neural network is retrained until the prediction accuracy reaches the set accuracy, thereby obtaining a stripe rust prediction model.
[0021] Optionally, the method for obtaining meteorological factors in the historical period specifically includes:
[0022] A correlation analysis is performed on the overall occurrence trend of wheat stripe rust in the historical period and the relevant meteorological factors in the corresponding historical period to obtain multiple correlation coefficients; the relevant meteorological factors include: temperature factor, precipitation factor, sunshine duration factor, snow accumulation factor and relative humidity factor;
[0023] The relevant meteorological factors corresponding to the correlation coefficients greater than the set coefficient value are determined as the meteorological factors within the historical period.
[0024] Optionally, the meteorological suitability of the disease in the future period of the year is determined based on the meteorological elements of each day in the future period of the year, specifically including:
[0025] Determine the comprehensive index of meteorological suitability for wheat stripe rust in the future period of the year according to the meteorological elements of each day in the future period of the year;
[0026] The meteorological suitability of the disease in the future period of the year is determined based on the comprehensive meteorological suitability index of wheat stripe rust.
[0027] Optionally, the expression of the comprehensive index of meteorological suitability for wheat stripe rust is:
[0028] Z=0.25Y T +0.3Y R +0.35Y P +0.1Y S ;
[0029] Where Z is the comprehensive index of meteorological suitability for wheat stripe rust, Y T is the temperature suitability, Y R is the relative humidity suitability, Y P is the precipitation suitability, Y S It is the suitability of sunlight.
[0030] Optionally, the expression of the temperature suitability is:
[0031]
[0032] Where T is the average temperature in the future period.
[0033] Optionally, the expression of relative humidity suitability is:
[0034]
[0035] Where R is the average relative humidity in the future period.
[0036] Optionally, the expression of precipitation suitability is:
[0037]
[0038] Where P is the accumulated precipitation in the future period.
[0039] Optionally, the expression of sunshine suitability is:
[0040]
[0041] Where S is the cumulative sunshine hours in the future period.
[0042] Optionally, the target meteorological factors include: the average temperature in December of the previous year, the average minimum temperature in December of the previous year, the average temperature in February of the current year, the average temperature in winter, the average minimum temperature in winter, the precipitation in January of the current year, the precipitation in winter and the sunshine hours in February of the current year.
[0043] According to the specific embodiments provided in this application, this application has the following technical effects:
[0044] The present application provides a method for predicting wheat stripe rust. The wheat greening and rising period is the early spring season when wheat stripe rust begins to occur. The overall occurrence trend of wheat stripe rust in the current year is output through the stripe rust prediction model to predict the degree of potential occurrence trend of stripe rust in the current year, thereby realizing the long-term trend forecast of stripe rust. During the high incidence period of the spread and epidemic of stripe rust, the meteorological suitability of the disease is determined by the daily meteorological conditions in the future period, so as to judge whether the future period is suitable for the spread and development of wheat stripe rust, thereby realizing the daily dynamic forecast of the development of stripe rust in the future period. The present application predicts wheat stripe rust in stages by combining the methods under the above two stages, taking into account the large-scale development trend of the pathogen after overwintering under winter meteorological conditions, and taking into account the small-scale development trend of whether the meteorological conditions in the epidemic stage are suitable for its spread and epidemic, to predict the probability of occurrence of wheat stripe rust in the future period of the current year, thereby realizing the combined forecast of long-term and short-term forecasts. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0046] Figure 1 Schematic diagram of a process for predicting wheat stripe rust in one embodiment of the present application. DETAILED DESCRIPTION
[0047] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0048] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0049] This application mainly adopts the combination of BP neural network + disease meteorological suitability to achieve long-term and short-term wheat stripe rust meteorological risk level forecast. The specific implementation example is as follows:
[0050] In an exemplary embodiment, Figure 1 As shown, a method for predicting wheat stripe rust is provided, comprising the following steps S1 to S5, wherein:
[0051] S1: During the wheat greening and sprouting period, the target meteorological factors are obtained; the target meteorological factors include the meteorological factors of the current year and the meteorological factors of the previous year.
[0052] Among them, meteorological factors are the key meteorological factors that affect the occurrence of wheat stripe rust in spring of that year. Based on the screening of key meteorological factors such as light, temperature and water that are closely related to the occurrence of wheat stripe rust, a stripe rust prediction model was constructed using BP neural network to forecast the meteorological risk level of wheat stripe rust and achieve long-term forecast of the potential occurrence trend of stripe rust.
[0053] S2: Input the target meteorological factors into the stripe rust prediction model to predict the overall occurrence trend of wheat stripe rust in that year.
[0054] Among them, the stripe rust prediction model is obtained by iterative training of the BP neural network model; the overall occurrence trend of wheat stripe rust is used to characterize the potential occurrence of wheat stripe rust during the wheat greening and rising period.
[0055] The overall trend of wheat stripe rust is determined according to the Technical Specifications for Wheat Stripe Rust Monitoring and Forecasting (GB / T15795-2011). It is divided into 5 levels according to the disease index or diseased field rate index. Specifically, the overall trend of wheat stripe rust in the current year is expressed in the occurrence level (1, 2, 3, 4, 5 levels). The wheat greening and rising period is March. From mid-to-late February, it is the early spring season when wheat stripe rust begins to occur. The specific time is March 1st to predict the general trend of the potential occurrence and development of stripe rust in the current year.
[0056] S3: During the high incidence period of wheat diseases, obtain the meteorological elements for each day in the future period of the year.
[0057] Meteorological factors include temperature, precipitation, relative humidity and sunshine hours. The future period is within 3 days from the first day of the high incidence period of diseases; the high incidence period of wheat diseases is April.
[0058] S4: Determine the meteorological suitability of the disease in the future period of the year based on the daily meteorological elements in the future period of the year.
[0059] Among them, the meteorological suitability of diseases is used to characterize whether the meteorological conditions in the future period are suitable for the spread and development of wheat stripe rust. The meteorological suitability of diseases refers to the degree of favorable influence of meteorological factors such as temperature, precipitation, relative humidity and sunshine on the key stages of the occurrence and development of crop pathogenic microorganisms, such as reproduction, spread and infection.
[0060] The grading index of the meteorological suitability forecast for diseases is based on the national meteorological industry standard "Method for Forecasting Meteorological Grades of Crop Diseases", which is divided into 4 levels according to the meteorological suitability of diseases. Specifically, the meteorological suitability of diseases is expressed in 4 levels: very suitable, suitable, sub-suitable and unsuitable. The period from the end of March to the heading period of wheat in April is the peak period for the spread of stripe rust. The specific time is April 1st to judge whether it is suitable for the spread and development of stripe rust, so as to realize the daily dynamic forecast of the development of stripe rust in the future period.
[0061] S5: The prediction results are obtained based on the overall occurrence trend of wheat stripe rust in that year and the meteorological suitability of the disease.
[0062] Among them, the prediction results include the probability of occurrence of wheat stripe rust in the future period of the year.
[0063] In another exemplary embodiment of the present application, a method for determining a stripe rust prediction model specifically includes:
[0064] Obtain training data; the training data includes: meteorological factors in a historical period and the corresponding overall occurrence trend of wheat stripe rust.
[0065] The training data is divided into training learning samples and prediction samples.
[0066] The training learning samples are input into the BP neural network for iterative training. When the set number of iterations is reached, the trained BP neural network model is obtained.
[0067] The trained BP neural network model was verified according to the forecast samples to obtain the forecast accuracy; the forecast accuracy was obtained by comparing the overall occurrence trend of wheat stripe rust in the forecast samples with the overall occurrence trend of wheat stripe rust output by the trained BP neural network model.
[0068] If the prediction accuracy reaches the set accuracy, the trained BP neural network model is determined as the stripe rust prediction model.
[0069] If the prediction accuracy does not reach the set accuracy, the number of training learning samples is increased, and the BP neural network is retrained until the prediction accuracy reaches the set accuracy, thereby obtaining a stripe rust prediction model.
[0070] In another exemplary embodiment of the present application, a method for obtaining meteorological factors in a historical period specifically includes:
[0071] A correlation analysis was conducted on the overall occurrence trend of wheat stripe rust in the historical period and the relevant meteorological factors in the corresponding historical period, and multiple correlation coefficients were obtained; the relevant meteorological factors included: temperature factor, precipitation factor, sunshine duration factor, snow accumulation factor and relative humidity factor.
[0072] The relevant meteorological factors corresponding to the correlation coefficients greater than the set coefficient value are determined as the meteorological factors within the historical period.
[0073] As an optional implementation, the overall occurrence trend of wheat stripe rust from 2006 to 2024 was analyzed with 44 relevant meteorological factors (22 temperature factors, 8 precipitation factors, 7 sunshine hours factors, 5 snow depth factors, and 2 relative humidity factors) including November and December of the previous year, January, February, and March of the current year, and the average temperature, minimum temperature, sunshine hours, precipitation, and maximum snow depth in winter. The results in Table 1 show that the overall occurrence trend of wheat stripe rust is significantly correlated with the average temperature in December of the previous year. The correlation coefficients of the average temperature in December of the previous year, the average temperature in February, and the precipitation in January were significantly positively correlated, and the correlation coefficients were 0.489, 0.544, 0.525, and 0.546, respectively (the correlation coefficients of significant positive correlation are marked with * in the table); and the correlation coefficients of the sunshine hours in February, the average temperature in winter, the average minimum temperature in winter, and the precipitation in winter were extremely significantly positively correlated, and the correlation coefficients were 0.654, 0.633, 0.640, and 0.599, respectively (the correlation coefficients of extremely significant positive correlation are marked with ** in the table). Therefore, it is determined that the eight factors of the average temperature in December of the previous year, the average minimum temperature in December, the average temperature in February, the average temperature in winter, the average minimum temperature in winter, the precipitation in January, the precipitation in winter, and the sunshine hours in February are the key meteorological factors affecting the occurrence and development of wheat stripe rust in spring.
[0074] Table 1 Relationship between the overall occurrence trend of wheat stripe rust and related meteorological factors
[0075]
[0076]
[0077] Taking 2006-2022 as the training sample and 2023-2024 as the forecast sample, eight meteorological factors that will significantly affect the occurrence of wheat stripe rust are (average temperature in December of the previous year T6, average minimum temperature in December of the previous year T8, average temperature in February T 14 , January precipitation R3, February sunshine hours S4, winter average temperature T 17 , the average minimum temperature in winter T 19 , winter precipitation R5) as the input layer of the BP neural network model, and the overall occurrence trend of wheat stripe rust as the output layer. When the training rate is 0.1 and the number of training times is 1000, the minimum fitting error is 0.000172, and the model has the best fitting degree.
[0078] As shown in Table 2, the overall accuracy of the stripe rust prediction model is 100% for the forecast sample of 2023-2024, and the model has good performance in predicting the overall occurrence trend of wheat stripe rust.
[0079] Table 2 Forecast evaluation of the overall occurrence trend of wheat stripe rust in 2023 and 2024
[0080]
[0081]
[0082] In another exemplary embodiment of the present application, determining the meteorological suitability of the disease in the future period of the year according to the meteorological elements of each day in the future period of the year specifically includes:
[0083] The comprehensive meteorological suitability index for wheat stripe rust in the future period of the year is determined based on the daily meteorological elements in the future period of the year.
[0084] The meteorological suitability of wheat stripe rust in the future period of the year is determined based on the comprehensive meteorological suitability index of wheat stripe rust.
[0085] In another exemplary embodiment of the present application, the expression of the comprehensive index of meteorological suitability for wheat stripe rust is:
[0086] Z=0.25Y T +0.3Y R +0.35Y P +0.1Y S .
[0087] Where Z is the comprehensive index of meteorological suitability for wheat stripe rust, Y T is the temperature suitability, Y R is the relative humidity suitability, Y P is the precipitation suitability, Y S It is the suitability of sunlight.
[0088] The higher the value of the comprehensive index of meteorological suitability for wheat stripe rust, the higher the suitability of the meteorological level for crop diseases. The index is used to determine whether the weather is suitable for the spread and development of stripe rust. The meteorological suitability for wheat stripe rust, i.e., the meteorological suitability for the disease, is divided into four levels: very suitable, suitable, sub-suitable, and unsuitable. The specific classification standards are shown in Table 3:
[0089] Table 3 Classification standard of comprehensive index of meteorological suitability for wheat stripe rust
[0090] Meteorological suitability for wheat stripe rust Comprehensive index of meteorological suitability Z Very suitable 0.7≤Z≤1 suitable 0.54≤Z＜0.7 Second appropriate 0.32≤Z＜0.54 Unsuitable Z＜0.32
[0091] In another exemplary embodiment of the present application, the expression of temperature suitability is:
[0092]
[0093] Where T is the average temperature in the future period, in degrees Celsius (℃).
[0094] In another exemplary embodiment of the present application, the expression of relative humidity suitability is:
[0095]
[0096] Where R is the average relative humidity in the future period, expressed as a percentage (%).
[0097] In another exemplary embodiment of the present application, the precipitation suitability Y P Calculation formula:
[0098]
[0099] Where P is the cumulative precipitation in the future period, in millimeters (mm).
[0100] In another exemplary embodiment of the present application, the sunshine suitability Y S Calculation formula:
[0101]
[0102] Where S is the cumulative sunshine hours in the future period, in hours (h).
[0103] In another exemplary embodiment of the present application, the target meteorological factors include: the average temperature in December of the previous year, the average minimum temperature in December of the previous year, the average temperature in February of the current year, the average temperature in winter, the average minimum temperature in winter, the precipitation in January of the current year, the precipitation in winter and the sunshine hours in February of the current year.
[0104] This method starts from the characteristics of pathogens' overwintering, diffusion and epidemic spread, and takes into account the large-scale development trend of pathogens' potential occurrence after overwintering under winter meteorological conditions, and also takes into account the small-scale development trend of whether meteorological conditions are suitable for its spread and epidemic spread during the epidemic stage, thus realizing the combined forecast of long-term and short-term forecasts. It has both long-term large trend forecasts and short-term nowcasts with high forecast accuracy, which not only provides sufficient preparation time for disease prevention and control, but also greatly improves the accuracy and timeliness of meteorological risk level forecasts.
[0105] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0106] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.
Claims
1. A method for predicting wheat stripe rust, characterized in that: The wheat stripe rust prediction method comprises: During the wheat greening and rooting period, target meteorological factors are obtained; the target meteorological factors include the meteorological factors of the current year and the meteorological factors of the previous year; the wheat greening and rooting period is March; Inputting the target meteorological factors into a stripe rust prediction model to predict the overall occurrence trend of wheat stripe rust in that year; the stripe rust prediction model is obtained by iteratively training a BP neural network; During the high incidence period of wheat diseases, obtain the meteorological elements of each day in the future period of the year; the meteorological elements include: temperature, precipitation, relative humidity and sunshine hours; the future period is within 3 days from the first day of the high incidence period of wheat diseases; the high incidence period of wheat diseases is April; Determine the meteorological suitability of the disease in the future period of the year according to the meteorological elements of each day in the future period of the year; the meteorological suitability of the disease is used to indicate whether the meteorological conditions in the future period are suitable for the spread and development of wheat stripe rust; The prediction result is obtained based on the overall occurrence trend of wheat stripe rust in the current year and the meteorological suitability of the disease; the prediction result includes the probability of occurrence of wheat stripe rust in the future period of the current year.
2. The wheat stripe rust prediction method according to claim 1, characterized in that: The method for determining the stripe rust prediction model specifically includes: Acquire training data; the training data includes: meteorological factors in a historical period and the corresponding overall occurrence trend of wheat stripe rust; Dividing the training data to obtain training learning samples and forecast samples; Inputting the training learning samples into the BP neural network for iterative training, and obtaining a trained BP neural network model when a set number of iterations is reached; The trained BP neural network model is verified according to the forecast sample to obtain the forecast accuracy; the forecast accuracy is obtained by comparing the overall occurrence trend of wheat stripe rust in the forecast sample with the overall occurrence trend of wheat stripe rust output by the trained BP neural network model; If the prediction accuracy reaches the set accuracy, the trained BP neural network model is determined as the stripe rust prediction model; If the prediction accuracy does not reach the set accuracy, the number of the training learning samples is increased, and the BP neural network is retrained until the prediction accuracy reaches the set accuracy, thereby obtaining a stripe rust prediction model.
3. The wheat stripe rust prediction method according to claim 2, characterized in that: The method for obtaining meteorological factors in the historical period includes: A correlation analysis is performed on the overall occurrence trend of wheat stripe rust in the historical period and the relevant meteorological factors in the corresponding historical period to obtain multiple correlation coefficients; the relevant meteorological factors include: temperature factor, precipitation factor, sunshine duration factor, snow accumulation factor and relative humidity factor; The relevant meteorological factors corresponding to the correlation coefficients greater than the set coefficient value are determined as the meteorological factors within the historical period.
4. The wheat stripe rust prediction method according to claim 1, characterized in that: The meteorological suitability of the disease in the future period of the year is determined based on the meteorological elements of each day in the future period of the year, including: Determine the comprehensive index of meteorological suitability for wheat stripe rust in the future period of the year according to the meteorological elements of each day in the future period of the year; The meteorological suitability of the disease in the future period of the year is determined based on the comprehensive meteorological suitability index of wheat stripe rust.
5. The wheat stripe rust prediction method according to claim 4, characterized in that: The expression of the comprehensive index of meteorological suitability for wheat stripe rust is: Z=0.25Y T +0.3Y R +0.35Y P +0.1Y S ; Where Z is the comprehensive index of meteorological suitability for wheat stripe rust, Y T is the temperature suitability, Y R is the relative humidity suitability, Y P is the precipitation suitability, Y S It is the suitability of sunlight.
6. The wheat stripe rust prediction method according to claim 5, characterized in that: The expression of the temperature suitability is: Where T is the average temperature in the future period.
7. The wheat stripe rust prediction method according to claim 5, characterized in that: The expression of relative humidity suitability is: Where R is the average relative humidity in the future period.
8. The method for predicting wheat stripe rust according to claim 5, characterized in that: The expression of precipitation suitability is: Where P is the accumulated precipitation in the future period.
9. The wheat stripe rust prediction method according to claim 5, characterized in that: The expression of sunshine suitability is: Where S is the cumulative sunshine hours in the future period.
10. The wheat stripe rust prediction method according to claim 1, characterized in that: The target meteorological factors include: the average temperature in December of the previous year, the average minimum temperature in December of the previous year, the average temperature in February of the current year, the average temperature in winter, the average minimum temperature in winter, the precipitation in January of the current year, the precipitation in winter and the sunshine hours in February of the current year.