Agricultural meteorological risk analysis and index insurance model construction method

Abstract

The invention relates to the technical field of agricultural insurance, in particular to an agricultural meteorological risk analysis and index insurance model construction method, which comprises the following steps: acquiring historical meteorological data, crop yield data and actual yield; determining a crop trend yield, and calculating a meteorological yield, a relative meteorological yield and a meteorological yield reduction rate; establishing an index disaster damage model and a per unit yield distribution model; determining a compensation structure by using a projection pursuit method; and calculating a pure premium rate and determining an actual premium rate, and finally outputting the premium. Therefore, the problems of low automation degree of agricultural meteorological index insurance design, low matching degree of insurance products and actual agricultural production risks, lack of scientificity and reasonability in insurance rate determination, insufficient accuracy and applicability of the insurance products and the like are solved, and a more effective risk guarantee mechanism is provided for farmers.

Description

Technical Field

[0001] This application relates to the technical field of agricultural insurance, and particularly relates to a method for analyzing agricultural meteorological risks and constructing an index insurance model. Background Art

[0002] Agricultural meteorological index insurance is an innovative agricultural insurance product based on meteorological indices. It quantifies the relationship between meteorological data and crop yield losses to provide risk protection for farmers. This insurance plays an important role in agricultural risk management, farmer income protection, and promoting sustainable agricultural development.

[0003] Existing design methods for agricultural meteorological index insurance mainly rely on manual experience and traditional statistical models. However, these methods have the following problems: Complex data processing: Agricultural meteorological data and crop yield data are usually large, complex, and multi-dimensional. Traditional methods are often time-consuming and laborious in processing these data, with low efficiency.

[0004] Imprecise index design: If the relationship between meteorological indices and crop yields is not clear or there are complex non-linear relationships, traditional index design methods are difficult to accurately capture these relationships, resulting in a reduction in the effectiveness and accuracy of insurance products.

[0005] Unreasonable claim settlement plan design: Existing claim settlement plan designs are often overly simplistic and difficult to reflect the actual impact of different degrees of meteorological disasters on crop yields, which may lead to unfair claims or excessive risks borne by insurance companies.

[0006] Lack of scientific basis for premium rating: Traditional premium rating methods mainly rely on historical data and empirical judgment, and are difficult to accurately reflect the current risk situation and future climate change trends, which may lead to unreasonable pricing of insurance products.

[0007] Insufficient adaptability: Different regions and different crops have different sensitivities to meteorological factors, and existing design methods are difficult to flexibly adapt to these differences, restricting the popularization and application of insurance products.

[0008] Low degree of automation: Most existing design processes require a large amount of manual intervention, which is not only inefficient but also prone to introducing subjective biases, affecting the scientific nature and fairness of insurance products.

[0009] In summary, these problems have led to low design quality of agricultural meteorological index insurance products and difficulty in meeting the needs of farmers and insurance companies. Summary of the Invention

[0010] This application provides a method for agricultural meteorological risk analysis and index insurance model construction to solve the problems in the prior art, such as the design of agricultural insurance products relying on manual experience, cumbersome data processing, difficulty in accurately quantifying the relationship between meteorological indices and crop yield reduction, lack of scientific basis for insurance rate determination, and inaccurate design of compensation plans.

[0011] In the first aspect of the embodiments of this application, a method for agricultural meteorological risk analysis and index insurance model construction is provided, including the following steps: obtaining historical meteorological data, crop yield data, and actual crop yield, where the historical meteorological data includes daily average temperature, daily maximum temperature, daily minimum temperature, daily precipitation, daily maximum wind speed, daily extreme wind speed, relative humidity, and sunshine hours; determining the trend yield of crops based on the historical meteorological data and crop yield data, determining the meteorological yield according to the trend yield of crops and the actual crop yield, calculating the relative meteorological yield according to the meteorological yield and the trend yield of crops, constructing a trend yield model using the target fitting method based on the historical meteorological data and the relative meteorological yield, and calculating the meteorological yield reduction rate according to the relative meteorological yield and the actual crop yield; using the regression analysis method to establish an index disaster loss model between the meteorological index and the meteorological yield reduction rate, and establishing a yield distribution model based on the index disaster loss model and the historical meteorological data; according to the yield distribution model and the index disaster loss model, segmenting the meteorological index according to a preset rule based on the projection pursuit method, calculating the insurance compensation ratio for each level of index, and determining the compensation structure; calculating the pure insurance rate based on the compensation structure, and determining the actual insurance rate according to the safety factor, operating expense coefficient, and profit margin, and outputting the insurance premium.

[0012] Preferably, before establishing the index disaster loss model between the meteorological index and the meteorological yield reduction rate, it further includes: calculating the historical values of multiple relevant indices based on the historical meteorological data, constructing an index set according to the historical values of the multiple relevant indices, where the index types in the index set include high temperature index, low temperature index, heat accumulation temperature index, cold accumulation index, heavy rain index, continuous days index, and wind disaster index; setting different index thresholds for multiple index types respectively, and calculating the corresponding index values according to the different index thresholds; determining the critical thresholds for triggering insurance compensation for multiple index types based on the historical meteorological data; comparing the index value with the critical threshold, and triggering insurance compensation when the index value exceeds the critical threshold, otherwise not starting the compensation mechanism.

[0013] Preferably, before establishing the index disaster loss model between the meteorological index and the meteorological yield reduction rate, it further includes: calculating the historical value sequence of each meteorological index based on the historical meteorological data, and determining the distribution characteristics of the historical value sequence; determining a reasonable threshold interval according to the distribution characteristics, and calculating the occurrence probability of the meteorological index threshold according to the threshold interval and the historical value sequence.

[0014] Preferably, an index disaster loss model between the meteorological index and the meteorological reduction rate is established, including: pairing the historical value sequence with the meteorological reduction rate data, and using the regression analysis method to establish a relationship model between the meteorological index and the meteorological reduction rate; inputting the meteorological index threshold into the relationship model to calculate the meteorological reduction rate; for the meteorological index, establishing an index disaster loss model including the index threshold, occurrence probability, and reduction rate; evaluating the goodness of fit of the index disaster loss model, and selecting the target model as the final index disaster loss model.

[0015] Preferably, based on the projection pursuit method, the meteorological index is segmented according to a preset rule: according to the historical meteorological data, the meteorological index range is divided into multiple intervals; for each interval, calculate its corresponding compensation ratio, and construct a piecewise linear compensation structure according to the compensation ratio; use the numerical optimization method to adjust the boundaries of each interval to minimize the fitting error between the piecewise linear compensation structure and the actual loss data.

[0016] Preferably, the formula of the compensation structure is:

[0017]

[0018] Where is the adjusted compensation structure, is the compensation ratio of the i-th segment, is the upper limit of the weather index of the i-th segment, is the lower limit of the weather index of the i-th segment, is the upper limit of the reduction rate corresponding to the weather index of the i-th segment, is the lower limit of the reduction rate corresponding to the weather index of the i-th segment, VH is the actual value of the weather index.

[0019] Preferably, the formula for determining the premium rate of the meteorological index insurance is: , Where is the expected value of the loss rate, is the crop reduction rate when the i-th type of disaster occurs, is the probability that this type of disaster may occur; The formula for the actual insurance premium rate is: R' = R * (1 + safety factor) * (1 + operating expense factor) * (1 + profit margin).

[0020] The second aspect of the embodiments of the present application provides an agricultural meteorological risk analysis and index insurance model construction device, including: an acquisition module, configured to acquire historical meteorological data, crop yield data, and actual crop yield, where the historical meteorological data includes daily average temperature, daily maximum temperature, daily minimum temperature, daily precipitation, daily maximum wind speed, daily extreme wind speed, relative humidity, and sunshine hours; a calculation module, configured to determine the trend yield of crops based on the historical meteorological data and crop yield data, determine the meteorological yield according to the trend yield of crops and the actual crop yield, calculate the relative meteorological yield according to the meteorological yield and the trend yield of crops, construct a trend yield model using a target fitting method based on the historical meteorological data and the relative meteorological yield, and calculate the meteorological yield reduction rate according to the relative meteorological yield and the actual crop yield; an establishment module, configured to establish an index disaster loss model between the meteorological index and the meteorological yield reduction rate using the regression analysis method, and establish a yield distribution model based on the index disaster loss model and the historical meteorological data; a determination module, configured to segment the meteorological index according to a preset rule based on the projection pursuit method according to the yield distribution model and the index disaster loss model, calculate the insurance compensation ratio of each level of index, and determine the compensation structure; an output module, configured to calculate the pure insurance rate based on the compensation structure, and determine the actual insurance rate according to the safety factor, operating expense coefficient, and profit margin, and output the insurance premium.

[0021] The third aspect of the embodiments of the present application provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the program to implement the agricultural meteorological risk analysis and index insurance model construction method as described in the above embodiments.

[0022] The fourth aspect of the embodiments of the present application provides a computer-readable storage medium, on which a computer program or instruction is stored, and when the computer program or instruction is executed, the agricultural meteorological risk analysis and index insurance model construction method as described in the above embodiments is implemented.

[0023] Therefore, the present application has at least the following beneficial effects: The embodiments of the present application obtain and process historical meteorological data and crop yield data through an automated method, and use various mathematical models and algorithms, such as the moving average method, curve fitting method, etc., to accurately determine the trend yield of crops and the meteorological yield reduction rate, improving the data processing efficiency and accuracy. By establishing an index disaster loss model between the meteorological index and the yield reduction rate, the precise quantification of the impact of meteorological factors on crop yield is realized, and the problem that it is difficult to accurately quantify the relationship between the meteorological index and crop yield reduction in traditional methods is solved.

[0024] This application designs the compensation structure using the projection pursuit method. By segmenting the meteorological index according to preset rules and calculating the insurance compensation ratio for each level of the index, the compensation plan becomes more scientific and reasonable, improving the accuracy and fairness of insurance products. At the same time, by establishing a yield distribution model and using various statistical methods, the scientific determination of insurance rates is achieved, overcoming the problem of lack of scientific basis in rate determination in traditional methods.

[0025] In addition, the method of this application can be flexibly applied to different types of crops and meteorological conditions in different regions, with strong adaptability and popularization value. Through the automated design process, the design efficiency of agricultural meteorological index insurance products is greatly improved, reducing human errors and providing a more reliable and effective insurance plan for insurance companies and farmers.

[0026] In summary, a method for agricultural meteorological risk analysis and index insurance model construction provided by this application not only improves the efficiency and accuracy of agricultural insurance product design, but also provides a more scientific and effective tool for agricultural risk management, which is of great significance for promoting the development of the agricultural insurance industry and improving the risk resistance ability of agricultural production.

[0027] The additional aspects and advantages of this application will be partly given in the following description, partly become obvious from the following description, or be understood through the practice of this application. Description of the Drawings

[0028] The above and / or additional aspects and advantages of this application will become obvious and easy to understand from the following description of the embodiments in conjunction with the drawings, where: Figure 1 is a flowchart of the method for agricultural meteorological risk analysis and index insurance model construction according to an embodiment of this application; Figure 2 is a flowchart of the meteorological impact analysis of crop yield according to an embodiment of this application; Figure 3 is a flowchart of the trend yield fitting method according to an embodiment of this application; Figure 4 is a schematic diagram of the meteorological yield reduction rate calculation process according to an embodiment of this application; Figure 5 is a flowchart of the index disaster loss model construction process according to an embodiment of this application; Figure 6 is a flowchart of the compensation structure design based on the projection pursuit method according to an embodiment of this application; Figure 7 is a flowchart of the insurance rate determination method according to an embodiment of this application; Figure 8Schematic diagram of the structure of an agricultural meteorological risk analysis and index insurance model construction device provided according to an embodiment of the present application; Figure 9 Schematic diagram of the structure of an electronic device provided according to an embodiment of the present application. Detailed implementation manners

[0029] The embodiments of the present application will be described in detail below. The examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present application, and should not be construed as a limitation to the present application.

[0030] A method for agricultural meteorological risk analysis and index insurance model construction according to an embodiment of the present application will be described below with reference to the accompanying drawings. In view of the problems existing in the design of agricultural insurance mentioned in the above background technology, the present application provides a method for agricultural meteorological risk analysis and index insurance model construction. In this method, by automatically processing historical meteorological data and crop yield data, a relationship model between meteorological indices and the reduction rate is established, and a scientific and reasonable compensation structure and insurance rate are designed, effectively improving the efficiency and accuracy of the design of agricultural insurance products, reducing human errors, and providing more reliable risk protection for farmers. Thus, the problems in the prior art that the design of agricultural insurance products relies on manual experience, the data processing is cumbersome, it is difficult to accurately quantify the relationship between meteorological indices and crop reduction, the determination of insurance rates lacks a scientific basis, and the design of compensation plans is not precise enough are solved.

[0031] Specifically, Figure 1 Schematic flow chart of the method for agricultural meteorological risk analysis and index insurance model construction provided by an embodiment of the present application.

[0032] As Figure 1 shown, the method includes the following steps: In step S101, historical meteorological data, crop yield data, and actual crop yields are obtained.

[0033] Among them, the historical meteorological data includes daily average temperature, daily maximum temperature, daily minimum temperature, daily precipitation, daily maximum wind speed, daily extreme wind speed, relative humidity, and sunshine hours.

[0034] It can be understood that the embodiments of the present application can provide a solid data foundation for subsequent model construction and analysis by obtaining comprehensive historical data, thereby improving the accuracy and reliability of the design of insurance products.

[0035] Specifically, the time span of the historical meteorological data is from January 1, 1981 to December 31, 2023. The crop yield data comes from the national statistical yearbook and the statistical yearbooks of each province, including the sown area, total output, and unit yield data of various crops at the provincial, municipal, and county levels over the years, and the statistical duration exceeds 10 years.

[0036] It should be noted that historical meteorological data usually comes from long-term observational records of meteorological stations and can comprehensively reflect the climate characteristics and change trends of a region. Crop yield data and actual yields reflect the growth conditions and harvest levels of crops.

[0037] In step S102, the trend yield of the crop is determined based on historical meteorological data and crop yield data. The meteorological yield is determined based on the crop trend yield and the actual crop yield. The relative meteorological yield is calculated based on the meteorological yield and the crop trend yield. A trend yield model is constructed using the target fitting method based on historical meteorological data and the relative meteorological yield. The meteorological reduction rate is calculated based on the relative meteorological yield and the actual crop yield.

[0038] Among them, the target fitting method can be the three-year moving average method, the five-year moving average method, the quadratic curve method, or the cubic curve method. Among them, the method with the best fitting effect is automatically selected as the final trend yield fitting method.

[0039] It can be understood that through the comparison and selection of multiple fitting methods in the embodiments of the present application, the long-term trend of crop yields can be captured more accurately, thereby better separating the impact of meteorological factors on yields.

[0040] Specifically, taking the three-year moving average method as an example, its calculation formula is as follows: First term: Y't = (5Y1 + 2Y2 - Y3) / 6; Middle term: Y't = (Yt-1 + Yt + Yt+1) / 3; Last term: Y't = (5Yn + 2Yn-1 - Yn-2) / 6; Among them, Y't is the trend yield and Yt is the actual yield.

[0041] It should be noted that automatically selecting the method with the best fitting effect includes: calculating the goodness-of-fit index for each of three or more methods (such as the three-year moving average method, the quadratic curve method, the cubic curve method). Among them, the goodness-of-fit index includes R² (coefficient of determination), mean squared error (MSE), or mean absolute error (MAE); when R² is the largest, MSE is less than a preset threshold (such as 0.05), and the difference in R² is less than a threshold (such as 0.02), then the method with the lowest computational complexity is selected; if there are multiple methods that simultaneously meet the conditions, the method with the highest correlation coefficient between the predicted value and the actual value sequence is taken as the target solution.

[0042] In step S103, using the regression analysis method, an exponential disaster loss model between the meteorological index and the meteorological reduction rate is established, and a single-yield distribution model is established based on the exponential disaster loss model and historical meteorological data.

[0043] Among them, the regression analysis method can include the stepwise regression method and the multiple regression analysis method.

[0044] It can be understood that by establishing an exponential disaster loss model in the embodiments of the present application, the impact degree of different meteorological indices on crop yield reduction can be quantified, providing a scientific basis for subsequent claim structure design and premium rate determination.

[0045] Specifically, the yield distribution model can adopt a normal distribution model, a Gamma distribution model, a Logistic distribution model or a three-parameter Weibull distribution model, and the optimal fitting model is selected through the AD test (Anderson-Darling test).

[0046] Taking the stepwise regression method as an example, its process includes: calculating the correlation between each independent variable and the dependent variable y and fitting the equation, selecting the independent variable with the largest correlation according to the correlation coefficient ranking, constructing a stepwise regression equation, calculating the standard deviation of each coefficient and ranking, and obtaining the order of the influence degree of meteorological indices.

[0047] In the embodiments of the present application, before establishing the exponential disaster loss model between the meteorological index and the meteorological yield reduction rate, it further includes: calculating the historical values of multiple correlation indices based on historical meteorological data, constructing an index set according to the historical values of the multiple correlation indices, wherein the index types in the index set include high temperature index, low temperature index, heat accumulation temperature index, cold accumulation index, rainstorm index, continuous days index and wind disaster index; setting different index thresholds for multiple index types respectively, and calculating the corresponding index values according to different index thresholds; determining the critical thresholds for triggering insurance claims for multiple index types based on historical meteorological data; comparing the index values with the critical thresholds, and triggering insurance claims when the index values exceed the critical thresholds, otherwise not starting the claim settlement mechanism.

[0048] Among them, different index types reflect different meteorological disaster risks. For example, the high temperature index reflects the heat damage risk that crops may suffer, and the rainstorm index reflects the possible flood risk, etc.

[0049] It can be understood that by constructing various types of meteorological indices in the embodiments of the present application, the impacts of various meteorological factors on crop yield can be comprehensively evaluated, so as to design more accurate and comprehensive insurance products.

[0050] Specifically, for the high temperature index, it can be defined as the number of times of events where the daily maximum temperature ≥ 30 °C during the insurance period, the threshold range is 30 - 40 °C, and each 1 °C is used as an index.

[0051] In the embodiments of the present application, establishing an index loss model between meteorological indices and meteorological yield reduction rates includes: pairing a historical value sequence with meteorological yield reduction rate data, and using the regression analysis method to establish a relationship model between meteorological indices and meteorological yield reduction rates; inputting meteorological index thresholds into the relationship model to calculate meteorological yield reduction rates; for meteorological indices, establishing an index loss model including index thresholds, occurrence probabilities, and yield reduction rates; evaluating the goodness of fit of the index loss model, and selecting the target model as the final index loss model.

[0052] It can be understood that in the embodiments of the present application, by pairing historical data with yield reduction rates, establishing a relationship model, and performing model evaluation and selection, the most suitable index loss model can be obtained, providing a reliable basis for subsequent insurance product design, effectively quantifying the relationship between meteorological indices and crop yield reduction, and making insurance products more in line with the actual situation.

[0053] Specifically, taking the high temperature index as an example, the high temperature index values of each year can be paired with the crop yield reduction rates of the corresponding years. Then, different regression models are tried, such as linear regression, quadratic regression, or exponential regression, etc. Assume that the quadratic regression model has the best fitting effect, and its equation is: y = ax² + bx + c, where y is the yield reduction rate and x is the high temperature index.

[0054] Substitute the previously determined high temperature index threshold into this equation to calculate the corresponding predicted yield reduction rate, and obtain a table including the high temperature index threshold, occurrence probability, and predicted yield reduction rate. This is the loss model of the high temperature index. At the same time, the coefficient of determination is used to evaluate the goodness of fit of the model, and the model with the highest R² is selected as the final index loss model.

[0055] In step S104, according to the yield per unit distribution model and the index loss model, the meteorological index is segmented according to a preset rule based on the projection pursuit method, the insurance compensation ratio of each level of index is calculated, and the compensation structure is determined.

[0056] Among them, the projection pursuit method is a mathematical optimization method used to find the optimal solution in a multi-dimensional space.

[0057] It can be understood that in the embodiments of the present application, by designing the compensation structure through the projection pursuit method, a more fair and reasonable compensation plan can be provided for farmers while ensuring the risk controllability of the insurance company.

[0058] In the embodiments of the present application, segmenting the meteorological index according to the preset rule based on the projection pursuit method is as follows: according to historical meteorological data, the meteorological index range is divided into multiple intervals; for each interval, calculate its corresponding compensation ratio, and construct a piecewise linear compensation structure according to the compensation ratio; use a numerical optimization method to adjust the boundaries of each interval to minimize the fitting error between the piecewise linear compensation structure and the actual loss data.

[0059] It is understandable that through the reasonable segmentation of meteorological indices and the use of numerical optimization techniques to adjust the segmentation boundaries, such a method can obtain a compensation structure that not only conforms to the actual loss situation but also balances the interests of insurance companies and farmers. This design method of the compensation structure can significantly improve the fairness and effectiveness of agricultural insurance products, making the compensation more accurate and at the same time controlling the risks of insurance companies.

[0060] Specifically, taking the high-temperature index as an example, the high-temperature index range is divided into 5 intervals, such as 0 - 10 days, 11 - 20 days, 21 - 30 days, 31 - 40 days, and more than 41 days. For each interval, calculate its corresponding average yield reduction rate according to the previously established index disaster loss model, and use this yield reduction rate as the compensation ratio for this interval. Then, construct a piecewise linear compensation structure. For example, when the high-temperature index is 25 days, the compensation ratio may be 30%. Use optimization algorithms such as gradient descent to fine-tune the boundaries of these intervals to minimize the error between this piecewise linear compensation structure and the historical actual loss data. Through this method, a more realistic compensation structure can be obtained.

[0061] In the embodiment of the present application, the formula for the compensation structure is:

[0062] where is the adjusted compensation structure, and the historical compensation rate in the recent n years can be obtained, then a = average compensation rate in the recent n years before adjustment / expected compensation rate.

[0063] In step S105, calculate the pure insurance rate based on the compensation structure, and determine the actual insurance rate according to the safety factor, operating expense factor, and profit margin, and output the insurance premium.

[0064] In the embodiment of the present application, the formula for the pure insurance rate is: , where is the expected value of the loss rate, is the crop yield reduction rate when the i-th type of disaster occurs, is the probability that this type of disaster may occur.

[0065] In the embodiment of the present application, the formula for the actual insurance rate is: R' = R * (1 + safety factor) * (1 + operating expense factor) * (1 + profit margin).

[0066] It is understandable that through the scientific premium rating method in the embodiment of the present application, it can ensure that the pricing of insurance products is reasonable, which can not only provide appropriate protection for farmers but also ensure the business sustainability of insurance companies.

[0067] Specifically, assume that the safety factor of corn insurance in a certain area is 0.1, the operating expense factor is 0.2, the profit rate is 0.05, and the pure insurance rate is 7%. Then the actual insurance rate is: 7% * (1 + 0.1) * (1 + 0.2) * (1 + 0.05) = 9.702%. If the agreed insurance amount per mu is 1,000 yuan, then the insurance premium per mu is 97.02 yuan.

[0068] A method for agricultural meteorological risk analysis and index insurance model construction proposed according to an embodiment of the present application, by automatically processing historical meteorological data and crop yield data, establishing a relationship model between meteorological indices and the reduction rate, designing a scientific and reasonable compensation structure and insurance rate, effectively improving the efficiency and accuracy of agricultural insurance product design, reducing human errors, and providing more reliable risk protection for farmers. Thus, it solves the problems in the prior art such as the design of agricultural insurance products relying on artificial experience, cumbersome data processing, difficult to accurately quantify the relationship between meteorological indices and crop reduction, lack of scientific basis for insurance rate determination, and inaccurate design of compensation plans.

[0069] The method for agricultural meteorological risk analysis and index insurance model construction will be elaborated through a specific embodiment below, as Figure 2 shown, including: Step 1: Obtain historical meteorological data, crop yield data, and actual crop yields.

[0070] Among them, the historical meteorological data includes daily average temperature, daily maximum temperature, daily minimum temperature, daily precipitation, daily maximum wind speed, daily extreme wind speed, relative humidity, and sunshine hours. The time span of the historical meteorological data is from January 1, 1981 to December 31, 2023. The crop yield data comes from the national statistical yearbook and provincial statistical yearbooks, including the sown area, total output, unit yield data, cultivated land area, total sown area of crops, agricultural natural disaster disaster data, etc. at the provincial, municipal, and county levels over the years. And the statistical data is cleaned, including mutual verification of data at all levels, processing of outliers and missing values. The statistical duration exceeds 10 years.

[0071] Step 2: Determine the crop trend yield based on the historical meteorological data and crop yield data.

[0072] As Figure 3 shown, first try four methods for fitting the trend yield: three-year moving average method, five-year moving average method, quadratic curve method, and cubic curve method. Taking the three-year moving average method as an example, its calculation formula is as follows: The first term: Y't = (5Y1 + 2Y2 - Y3) / 6; The middle term: Y't = (Yt - 1 + Yt + Yt + 1) / 3; Last term: Y't = (5Yn + 2Yn-1 - Yn-2) / 6; Among them, Y't is the trend yield, and Yt is the actual yield. By comparing the fitting effects of different methods, the optimal method for fitting the trend yield is selected.

[0073] Specifically, the calculation of the three-year moving average method is shown in Table 1 below.

[0074] Table 1 Calculation of the three-year moving average method

[0075] Taking the five-year moving average method as an example, its calculation formula is as follows: First term: Y't = (3Y1 + 2Y2 + Y3 - Y5) / 5; Second term: Y't = (4Y1 + 3Y2 + 2Y3 + Y4) / 10; Middle term: Y't = (Yt-2 + Yt-1 + Yt + Yt+1 + Yt+2) / 5; Second-to-last term: Y't = (Y(n-3) + 2Y(n-2) + 3Y(n-1) + 4Yn) / 10; Last term: Y't = (-Y(n-4) + Y(n-2) + 2Y(n-1) + 3Yn) / 5; Among them, Y't is the trend yield, and Yt is the actual yield.

[0076] Specifically, the calculation of the five-year moving average method is shown in Table 2 below.

[0077] Table 2 Calculation of the five-year moving average method

[0078] Among them, Y't is the trend yield, t is the time (year), and a, b, and c are undetermined coefficients. These coefficients can be determined by the least squares method to minimize the sum of the squares of the errors between the fitting curve and the actual data.

[0079] Step 3: Determine the meteorological yield based on the crop trend yield and the actual crop yield, and calculate the relative meteorological yield based on the meteorological yield and the crop trend yield.

[0080] As Figure 4 shown, the formula for calculating the meteorological yield is: Yw = Y - Yt; The formula for calculating the relative meteorological yield is: Y'w = Yw / Yt.

[0081] Among them, Y is the actual yield, Yt is the trend yield, and Yw is the meteorological yield.

[0082] Step 4: Calculate the meteorological yield reduction rate based on the relative meteorological yield and the actual crop yield.

[0083] Among them, the calculation formula for the reduction rate is: y = -Y'w * 100% (when Y'w < 0); y = 0 (when Y'w ≥ 0).

[0084] Step 5: Use the regression analysis method to establish an exponential disaster loss model between the meteorological index and the meteorological reduction rate. As Figure 5 shown, first calculate the historical values of multiple relevant indexes based on historical meteorological data to construct an index set. The index types include high temperature index, low temperature index, heat accumulation temperature index, cold accumulation index, rainstorm index, consecutive days index, and wind disaster index. Then, use the stepwise regression or multiple regression method to establish a relationship model between the meteorological index and the reduction rate.

[0085] Taking stepwise regression as an example, its model expression is: Y = β0 + β1X1 + β2X2 +... + βkXk + ε; Among them, Y is the reduction rate, X1, X2,..., Xn are different meteorological indexes, β0, β1,..., βn are regression coefficients, and ε is a random error term.

[0086] Taking multiple regression as an example, its model expression is: Y = β0 + β1X1 + β2X2 +... + βkXk + ε; Among them, Y is the reduction rate, X1, X2,..., Xn are different meteorological indexes, β0, β1,..., βn are regression coefficients, and ε is a random error term. The multiple regression model expression is the same as that of stepwise regression. In terms of the fitting method, multiple regression will retain all independent variables (meteorological indexes) in the model. By standardizing the coefficients (β1, β2,…, βn) output by the multiple regression equation of each index and the reduction rate (standardized coefficient = original coefficient / standard deviation), and then sorting the standardized coefficients, the influence degree order of the meteorological indexes is obtained.

[0087] It should be noted that the relationship between the meteorological index and the reduction rate is shown in Table 3 below.

[0088] Table 3 Relationship between meteorological index and reduction rate

[0089] The construction of the index set and the determination of the threshold are shown in Table 4 below.

[0090] Table 4 Construction of index set and determination of threshold

[0091] Step 6: Based on the exponential disaster loss model and historical meteorological data, establish a single yield distribution model. Select the optimal fitting model from the normal distribution model, Gamma distribution model, Logistic distribution model, and three-parameter Weibull distribution model, and use the AD test (Anderson-Darling test) for model evaluation.

[0092] Specifically, the formula for the normal distribution model is: Probability density function: ; Cumulative distribution function:

[0093] where is a random variable, is the mean, is the standard deviation.

[0094] The formula for the Gamma distribution model is: Probability density function: ; Cumulative distribution function:

[0095] where is a random variable, is the location parameter, is the scale parameter.

[0096] The formula for the Logistic distribution model is: Probability density function: ; Cumulative distribution function: ; where is a random variable, is the shape parameter, is the scale parameter, is the Gamma function.

[0097] The formula for the three-parameter Weibull distribution model is: Probability density function: ; Cumulative distribution function: ; where is a random variable, is the shape parameter, is the scale parameter Step 7: According to the single yield distribution model and the exponential disaster loss model, segment the meteorological index based on the projection pursuit method according to preset rules, calculate the insurance compensation ratio for each level of index, and determine the compensation structure.

[0098] For exampleFigure 6 As shown, the formula for the compensation structure is:

[0099]

[0100] Wherein, is the adjusted compensation structure, is the compensation ratio for the i-th segment, is the upper limit of the weather index for the i-th segment, is the lower limit of the weather index for the i-th segment, is the upper limit of the yield reduction rate corresponding to the weather index of the i-th segment, is the lower limit of the yield reduction rate corresponding to the weather index of the i-th segment, and VH is the actual value of the weather index.

[0101] Step Eight: Calculate the pure insurance rate based on the compensation structure, and determine the actual insurance rate according to the safety factor, operating expense factor, and profit margin, and output the insurance cost.

[0102] As Figure 7 shown, the formula for the pure insurance rate is:

[0103] Wherein, is the expected value of the loss ratio, is the crop yield reduction rate when the i-th type of disaster occurs, is the probability that this type of disaster may occur. The formula for the actual insurance rate is: R' = R * (1 + safety factor) * (1 + operating expense factor) * (1 + profit margin).

[0104] In summary, through the automated processing of historical meteorological and crop yield data, the present invention uses a variety of mathematical models and algorithms, including trend yield fitting, regression analysis, projection pursuit method, etc., to achieve the precise design of agricultural meteorological index insurance. This method not only improves the data processing efficiency but also realizes the precise quantification of the impact of meteorological factors on crop yield reduction. By establishing a scientific index disaster loss model and compensation structure, as well as a reasonable insurance rate determination method, the present invention significantly improves the accuracy, fairness, and applicability of agricultural insurance products. This automated design method greatly reduces human error, improves the design efficiency, and at the same time has strong adaptability and can be flexibly applied to different regions and crop types. Finally, the present invention provides a more reliable and effective risk protection mechanism for farmers, and at the same time provides a more scientific product design and pricing basis for insurance companies, which is of great significance for promoting the development of the agricultural insurance industry and improving the risk resistance ability of agricultural production.

[0105] Next, a device for agricultural meteorological risk analysis and index insurance model construction according to an embodiment of the present application will be described with reference to the accompanying drawings.

[0106] Figure 8 It is a block diagram of a device for agricultural meteorological risk analysis and index insurance model construction according to an embodiment of the present application.

[0107] As Figure 8 shown, the device 10 for agricultural meteorological risk analysis and index insurance model construction includes: an acquisition module 100, a calculation module 200, an establishment module 300, a determination module 400, and an output module 500.

[0108] Among them, the acquisition module 100 is used to acquire historical meteorological data, crop yield data, and actual crop yields. Among them, the historical meteorological data includes daily average temperature, daily maximum temperature, daily minimum temperature, daily precipitation, daily maximum wind speed, daily extreme wind speed, relative humidity, and sunshine hours; the calculation module 200 is used to determine the trend yield of crops based on historical meteorological data and crop yield data, determine the meteorological yield according to the trend yield of crops and the actual crop yields, calculate the relative meteorological yield according to the meteorological yield and the trend yield of crops, construct a trend yield model using the target fitting method according to historical meteorological data and relative meteorological yield, and calculate the meteorological yield reduction rate according to relative meteorological yield and actual crop yields; the establishment module 300 is used to establish an index loss model between meteorological indices and meteorological yield reduction rates using the regression analysis method, and establish a yield distribution model based on the index loss model and historical meteorological data; the determination module 400 is used to segment the meteorological indices according to a preset rule based on the yield distribution model and the index loss model using the projection pursuit method, calculate the insurance compensation ratio of each level of index, and determine the compensation structure; the output module 500 is used to calculate the pure insurance rate based on the compensation structure, and determine the actual insurance rate according to the safety factor, operating expense factor, and profit margin, and output the insurance premium.

[0109] It should be noted that the foregoing explanation of the embodiment of the agricultural meteorological risk analysis and index insurance model construction method also applies to the device for agricultural meteorological risk analysis and index insurance model construction in this embodiment, and will not be elaborated here.

[0110] The agricultural meteorological risk analysis and index insurance model construction device proposed according to the embodiments of the present application can accurately quantify risks under complex meteorological conditions by intelligently processing historical meteorological and crop yield data, optimize insurance design under conventional climate conditions, automatically adjust the compensation structure and rate once an abnormal meteorological event is detected, effectively reducing the pricing deviation risk of agricultural insurance in extreme weather or climate change situations, ensuring the accuracy and applicability of insurance products, avoiding problems such as losses of insurance companies and insufficient protection for farmers caused by improper product design, and improving the reliability of agricultural meteorological index insurance and farmer satisfaction. Thus, it solves the problems in the prior art that when extreme climate events occur frequently or climate patterns change, the design of insurance products lags behind or even goes awry, resulting in poor protection effects and low acceptance by farmers, leading to operating risks of insurance companies and income losses of farmers.

[0111] Figure 9 The following is a schematic structural diagram of the electronic device provided by the embodiments of the present application. The electronic device may include: A memory 901, a processor 902, and a computer program stored on the memory 901 and executable on the processor 902.

[0112] When the processor 902 executes the program, it implements the agricultural meteorological risk analysis and index insurance model construction method provided in the above embodiments.

[0113] Further, the electronic device further includes: A communication interface 903 for communication between the memory 901 and the processor 902.

[0114] The memory 901 is used to store a computer program executable on the processor 902.

[0115] The memory 901 may include a high-speed RAM (Random Access Memory) memory, and may also include a non-volatile memory, such as at least one disk memory.

[0116] If the memory 901, the processor 902, and the communication interface 903 are implemented independently, the communication interface 903, the memory 901, and the processor 902 may be interconnected through a bus and communicate with each other. The bus may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 9It is represented by only one thick line, but it does not mean that there is only one bus or one type of bus.

[0117] Optionally, in a specific implementation, if the memory 901, the processor 902, and the communication interface 903 are integrated on a single chip, the memory 901, the processor 902, and the communication interface 903 can communicate with each other through an internal interface.

[0118] The processor 902 may be a CPU (Central Processing Unit), or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0119] The embodiments of the present application further provide a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned agricultural meteorological risk analysis and index insurance model construction method is implemented.

[0120] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0121] In addition, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present application, "a plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0122] Any process or method description depicted in the flowchart or otherwise described herein can be understood to represent a module, segment, or portion of code including one or more executable instructions for implementing a customized logical function or process. The scope of the preferred embodiments of the present application includes additional implementations, where the functions may be executed in a substantially simultaneous manner or in a reverse order according to the involved functions, rather than in the order shown or discussed. This should be understood by those skilled in the art to which the embodiments of the present application pertain.

[0123] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits with logic gate circuits for implementing logical functions on data signals, application specific integrated circuits with suitable combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), and the like.

[0124] Those of ordinary skill in the art of this technology can understand that all or part of the steps carried out in implementing the methods of the above embodiments can be completed by instructing relevant hardware through a program. The said program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0125] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A method for agricultural meteorological risk analysis and index insurance model construction, characterized in that: The method comprises: Acquiring historical meteorological data, crop yield data and actual crop yield, wherein the historical meteorological data includes daily average temperature, daily maximum temperature, daily minimum temperature, daily precipitation, daily maximum wind speed, daily maximum wind speed, relative humidity and sunshine hours; Determine the crop trend yield based on the historical meteorological data and the crop yield data, determine the meteorological yield according to the crop trend yield and the actual crop yield, calculate the relative meteorological yield according to the meteorological yield and the crop trend yield, construct a trend yield model using a target fitting method according to the historical meteorological data and the relative meteorological yield, and calculate the meteorological yield reduction rate according to the relative meteorological yield and the actual crop yield; Using regression analysis method, an exponential disaster loss model between the meteorological index and the meteorological yield reduction rate is established, and based on the exponential disaster loss model and the historical meteorological data, a yield distribution model is established; According to the yield distribution model and the index disaster loss model, the meteorological index is segmented according to preset rules based on the projection pursuit method, the insurance compensation ratio of each level of index is calculated, and the compensation structure is determined; The pure insurance premium rate is calculated based on the compensation structure, and the actual insurance premium rate is determined based on the safety factor, operating expense factor and profit margin, and the insurance cost is output.

2. The method for agricultural meteorological risk analysis and index insurance model construction according to claim 1, characterized in that: Before establishing the index disaster loss model between the meteorological index and the meteorological production reduction rate, the method includes: Calculating historical values ​​of multiple related indexes based on the historical meteorological data, and constructing an index set according to the historical values ​​of the multiple related indexes, wherein the index types in the index set include high temperature index, low temperature index, heat accumulation temperature index, accumulated cold index, rainstorm index, duration days index and wind disaster index; respectively setting different index thresholds for the plurality of index types, and calculating corresponding index values ​​according to the different index thresholds; Determining critical thresholds for triggering insurance payouts for a plurality of said index types based on said historical meteorological data; The index value is compared with the critical threshold value, and when the index value exceeds the critical threshold value, insurance compensation is triggered, otherwise the compensation mechanism is not activated.

3. The method for agricultural meteorological risk analysis and index insurance model construction according to claim 1, characterized in that: Before establishing the index disaster loss model between the meteorological index and the meteorological production reduction rate, it also includes: Calculating a historical value sequence of each meteorological index based on the historical meteorological data, and determining a distribution feature of the historical value sequence according to the historical value sequence; A reasonable threshold interval is determined according to the distribution characteristics, and the occurrence probability of the meteorological index threshold is calculated according to the threshold interval and the historical value sequence.

4. The method for agricultural meteorological risk analysis and index insurance model construction according to claim 3, characterized in that: An exponential disaster loss model between the meteorological index and the meteorological production reduction rate is established, including: Pairing the historical value sequence with the meteorological production reduction rate data, and using a regression analysis method to establish a relationship model between the meteorological index and the meteorological production reduction rate; Inputting the meteorological index threshold into the relationship model to calculate the meteorological production reduction rate; For the meteorological index, establishing an index disaster loss model including the index threshold, the occurrence probability and the production reduction rate; The goodness of fit of the exponential disaster loss model is evaluated, and the target model is selected as the final exponential disaster loss model.

5. The method for agricultural meteorological risk analysis and index insurance model construction according to claim 1, characterized in that: Based on the projection pursuit method, the meteorological index is segmented according to the preset rules: Dividing the meteorological index range into a plurality of intervals according to the historical meteorological data; For each of the intervals, the corresponding compensation ratio is calculated, and a piecewise linear compensation structure is constructed according to the compensation ratio; The boundaries of each interval are adjusted using a numerical optimization method to minimize the fitting error between the piecewise linear compensation structure and the actual loss data.

6. The method for agricultural meteorological risk analysis and index insurance model construction according to claim 5, characterized in that: The formula for the compensation structure is: ； ； in, To adjust the compensation structure, is the compensation ratio of the i-th segment, is the upper limit of the weather index in the i-th segment, is the lower limit of the weather index in the i-th section, is the upper limit of the reduction rate corresponding to the weather index in the i-th period, is the lower limit of the reduction rate corresponding to the weather index in the i-th period, V H is the actual value of the weather index.

7. The method for agricultural meteorological risk analysis and index insurance model construction according to claim 1, characterized in that: The formula for determining the weather index insurance premium rate is: ， in, is the expected value of the loss rate, is the crop yield reduction rate when the i-th disaster occurs, The probability of such disaster occurring; The formula for the actual insurance premium rate is: R'=R*(1+safety factor)*(1+operating expense coefficient)*(1+profit margin).

Images