Intelligent forecast method for the onset of plum rain in a region based on circulation adjustment and factor judgment

Abstract

The present invention relates to the technical field of intelligent forecasting of the onset of plum rain, and discloses a method for intelligent forecasting of the onset of plum rain in a region based on circulation adjustment and element judgment. The key points of the technical solution are as follows: determining multiple key characteristic quantities that affect the onset of plum rain, and calculating the membership of each characteristic quantity based on historical data; constructing a plum rain prediction credibility index; training a plum rain classification prediction model based on a long sequence historical data set using a support vector machine, a decision tree, or a logistic regression algorithm; acquiring circulation element data from a global forecast system in real time, extracting key characteristic quantities, calculating their membership, and inputting them into a prediction model to classify and judge the onset of plum rain conditions; and combining actual ground element monitoring and forecast data to determine whether plum rain monitoring standards are met, and comprehensively determining the onset date of the plum rain.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent forecasting of the onset of the plum rain season, and more specifically, to an intelligent forecasting method for the onset of the plum rain season in a region based on circulation adjustment and element judgment. Background Art

[0002] Traditional plum rain forecasting methods are highly subjective, require a large amount of data, and involve cumbersome analysis processes for forecasters. Forecasting requires considerable effort, and the judgment results are subjective and imprecise. Furthermore, traditional plum rain forecasting lacks structural logic and a clear hierarchy. It also fails to integrate live monitoring with forecast data, and lacks a seamless integration of circulation adjustments and other factors.

[0003] Therefore, the present invention provides an intelligent forecasting method for the onset of the plum rain season in a region based on circulation adjustment and factor judgment, which improves the above-mentioned technical problems. Summary of the Invention

[0004] The embodiments of the present disclosure aim to address the deficiencies of the existing technology and provide an intelligent forecasting method for the onset of the plum rain season in a region based on circulation adjustment and factor judgment. The intelligent forecasting method for the onset of the plum rain season in a region based on circulation adjustment and factor judgment combines characteristic quantities with hierarchical analysis and intelligent identification to complete an intelligent, accurate and practical forecasting method for the onset of the plum rain season, which is used to forecast the specific start time of the plum rain season in the Yangtze and Huaihe River basins every year.

[0005] The above technical objectives of the present invention are achieved through the following technical solutions: a method for intelligently forecasting the onset of the plum rain season in a region based on circulation adjustment and factor judgment, comprising the following steps:

[0006] S1. Determine multiple key characteristic quantities that affect the onset of plum rain, and calculate the membership degree of each characteristic quantity based on historical data;

[0007] S2. Constructing a credibility index for the prediction of the onset of the plum rain season. The credibility index is obtained by weighted summation of the membership of each characteristic quantity, and the weight distribution is determined by the hierarchical analysis method;

[0008] S3. Based on a long-sequence historical data set, a support vector machine, decision tree, or logistic regression algorithm is used to train a plum rain classification prediction model, wherein the input of the model includes the credibility index and the key feature quantity membership;

[0009] S4. Real-time acquisition of circulation element data from the global forecast system, extraction of key characteristic quantities and calculation of their membership, and input into the forecast model for classification and judgment of plum rain conditions;

[0010] S5. Combine the actual monitoring and forecast data of ground elements to determine whether the plum rain monitoring standards are met, including the regional precipitation threshold, the proportion of rainy days and the average temperature conditions, and comprehensively determine the date of the start of the plum rain season.

[0011] As a preferred technical solution of the present invention, multiple key characteristic quantities that affect the onset of plum rain include: 500hPa Kolkata wind direction, subtropical high pressure ridge, South Asian high pressure ridge, westerly jet stream, northeastern cold vortex, 850hPa pseudo-equivalent potential temperature, 850hPa low-level jet stream, 850hPa humidity, and 500hPa-8℃ line.

[0012] As a preferred technical solution of the present invention, a method for determining the position of the subtropical high pressure ridge line includes: identifying the subtropical high pressure center within the range of 10-60°N and 120°E longitude, and determining the latitude value when its zonal wind u=0 and du / dy>0;

[0013] The method for determining the position of the South Asian high pressure ridge includes: calculating the average of the latitude values ​​at different longitudes within the range of 1256 potential decameter lines in the range of 10-60°N and 100-120°E, which satisfies the zonal wind u=0 and du / dy>0.

[0014] As a preferred technical solution of the present invention, the membership assignment rule is: the membership of the feature quantity that meets the conditions for the arrival of the plum rain season is 1, the membership of the feature quantity that is close to the conditions for the arrival of the plum rain season is 0, and the membership of the feature quantity that does not meet the conditions for the arrival of the plum rain season is -1.

[0015] As a preferred technical solution of the present invention, the process of weight allocation using hierarchical analysis method is as follows: construct a judgment matrix through expert experience, calculate the weight value of each feature value, and perform consistency test. When the consistency ratio is less than 0.1, the weight allocation is determined to be valid.

[0016] As a preferred technical solution of the present invention, the support vector machine adopts a radial basis function kernel or a polynomial kernel, and the formula of the polynomial kernel is: ;in, and represents the input vector, represents the dot product of these two vectors, and d is the degree of the polynomial kernel.

[0017] As a preferred technical solution of the present invention, the decision tree adopts the CART algorithm, with the Gini index as the splitting criterion. The calculation formula of the Gini index is:

[0018] ;

[0019] in, Represents the proportion of category k in data set D, D represents the current data set, A represents a feature, and is used to split data set D into two subsets and , represents the total number of samples in the dataset D, and Represent subsets respectively and The number of samples in the Gini coefficient ( ) and the Gini coefficient ( ) represent subsets respectively and The Gini index ranges from [0, 1], and the smaller the value, the higher the purity of the dataset.

[0020] As a preferred technical solution of the present invention, the circulation element data obtained in real time include: ERA5 reanalysis data, and the forecast data are matched to the monitoring sites based on the grid interpolation method.

[0021] As a preferred technical solution of the present invention, the plum rain monitoring standard further includes:

[0022] a. More than one-third of the stations in the monitoring area have daily precipitation ≥ 0.1 mm, and the regional average daily precipitation is ≥ 2.0 mm;

[0023] b. The proportion of rainy days within 10 days from the first rainy day is ≥50%;

[0024] c. Daily average temperature ≥ 22℃.

[0025] As a preferred technical solution of the present invention, the rule for comprehensively determining the date of the beginning of the plum rain season is: if the circulation adjustment day is earlier than the day when the ground element conditions are met, the day when the ground element conditions are met will be the date of the beginning of the plum rain season; if the two dates are the same, the circulation adjustment day will be the date of the beginning of the plum rain season.

[0026] In summary, the present invention has the following beneficial effects:

[0027] First, by comprehensively screening nine key circulation characteristics, including the subtropical high-pressure ridge, the South Asian high-pressure ridge, and the 500hPa-8°C line, and combining them with surface precipitation, the proportion of rainy days, and temperature, the method quantifies the onset of the plum rain season in multiple dimensions. This effectively avoids the limitations of a single indicator and significantly improves forecast accuracy. Examples show that the predicted onset of the plum rain season in the Yangtze and Huaihe River basins in 2024 is fully consistent with actual monitoring results, validating the reliability of the method.

[0028] Second, a classification prediction model is constructed using machine learning algorithms such as support vector machines and decision trees. This model can automatically process massive amounts of historical and real-time meteorological data, reducing the subjective errors inherent in human judgment. Through model training and integrated optimization, intelligent identification of plum rain conditions is achieved, significantly improving forecast efficiency. Furthermore, the concept of fuzzy membership is introduced to convert qualitative evaluations (such as "approaching plum rain conditions") into quantitative scores (with memberships of 1, 0, and -1). Combined with the analytic hierarchy process (AHP), dynamic weights are assigned to feature variables, making the calculation of the credibility index (T) more scientific and rational, avoiding the drawback of traditional methods that rely on equally weighted addition of feature variables.

[0029] Third, real-time circulation data from global forecast systems (such as the ERA5 reanalysis) is integrated and matched to monitoring stations using gridded interpolation technology, achieving seamless integration of forecast data with real-time monitoring. Combined with model precipitation forecasts for the next week, this can provide early warning 7-10 days before the onset of the plum rain season, providing critical decision-making support for flood control and agricultural planning in the Yangtze and Huaihe River basins.

[0030] Fourth: Use the analytic hierarchy process to hierarchically judge the circulation adjustment indicators (such as the jump of the subtropical high ridge line and the northward movement of the westerly jet stream) and the ground element conditions (the proportion of rainy days and the temperature threshold), clarify the priority rules (circulation adjustment is a necessary condition, and ground elements are sufficient conditions), avoid indicator conflicts, and enhance the logical rigor of the forecast process. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A flowchart of a method for intelligently forecasting the onset of the plum rain season in a region based on circulation adjustment and factor judgment provided by an embodiment of the present invention;

[0032] Figure 2 A data architecture diagram provided for an embodiment of the present invention;

[0033] Figure 3 A diagram of the training framework for the plum rain classification prediction model provided in an embodiment of the present invention;

[0034] Figure 4 A schematic diagram of constructing a long sequence historical data set provided by an embodiment of the present invention;

[0035] Figure 5 A schematic diagram of the support vector machine algorithm provided in an embodiment of the present invention;

[0036] Figure 6 This is a flow chart for determining circulation adjustment indicators provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0037] The present application is described in detail below with reference to specific embodiments. The following embodiments will help those skilled in the art to further understand the present application, but are not intended to limit the present application in any form. It should be noted that those skilled in the art may make several variations and improvements without departing from the scope of the present application. These all fall within the scope of protection of the present application.

[0038] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0039] It should be noted that, if there is no conflict, the various features in the embodiments of the present application can be combined with each other and are all within the scope of protection of the present application. In addition, although the functional modules are divided in the device schematic and the logical order is shown in the flow chart, in some cases, the steps shown or described can be performed in a different order than the module division in the device or the order in the flow chart. In addition, the words "first", "second", "third", etc. used herein do not limit the data and execution order, but only distinguish between the same items or similar items with basically the same functions and effects.

[0040] Unless otherwise defined, all technical and scientific terms used in this specification have the same meanings as those commonly understood by those skilled in the art to which this application belongs. The terms used in this specification and in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application. The term "and / or" as used in this specification includes any and all combinations of one or more of the relevant listed items.

[0041] In addition, the technical features involved in each embodiment of the present application described below can be combined with each other as long as they do not conflict with each other.

[0042] The embodiment of the present disclosure aims to solve the problem of comprehensive analysis and judgment of the onset of plum rain based on a large amount of data. In view of this, the embodiment of the present disclosure proposes an intelligent forecast method for the onset of plum rain in a region based on circulation adjustment and factor judgment, combining characteristic quantities with hierarchical analysis and intelligent identification to complete an intelligent, accurate and practical forecast method for the onset of plum rain for the Yangtze and Huaihe River Basin during the flood season each year. The method combines the previous real-time monitoring of circulation and ground elements with future weather situation changes and factor forecasts, comprehensively considering the circulation situation adjustment, the real-time monitoring of specific weather conditions in different regions and plum rain monitoring standards, and using hierarchical analysis methods and intelligent identification and judgment to predict the onset of plum rain in advance based on the indicators of the onset of plum rain in the Yangtze and Huaihe River Basin.

[0043] Please refer to Figure 1 , Figure 1The flowchart of the method for intelligently forecasting the onset of plum rains in a region based on circulation adjustment and factor judgment according to an embodiment of the present disclosure is shown. The overall process mainly includes the following five steps:

[0044] Step 1: Determine multiple key characteristic quantities that affect the onset of plum rain and calculate the membership degree of each characteristic quantity based on historical data.

[0045] S1.1. Several key characteristics that affect the onset of the plum rain season include: the onset of the Indian monsoon (wind direction at Kolkata at 500hPa), the subtropical high-pressure ridge, the South Asian high-pressure ridge, the westerly jet stream, the northeastern cold vortex, the pseudo-equivalent potential temperature at 850hPa, the low-level jet stream at 850hPa, the humidity at 850hPa, and the -8°C line at 500hPa.

[0046] S1.2. Collect historical data on factors related to plum rain, and through comparison and synthesis analysis before and after the onset of plum rain, determine the quantified characteristic values ​​of each characteristic quantity before and after the onset of plum rain and the threshold of ground-truth monitoring elements, and determine the distribution spatial range of the calculated values ​​of each characteristic quantity. As shown in Table 1:

[0047] Table 1 Historical distribution of key circulation characteristics:

[0048] ;

[0049] Table 1 continued:

[0050] ;

[0051] S1.3. Convert the qualitative evaluation of whether the meteorological characteristics for the onset of the plum rain season are met into a quantitative evaluation.

[0052] Based on the historical data of meteorological circulation and elements on the day of the onset of the plum rain season and the 10 days before and after, key characteristic quantities were extracted and their membership degrees were calculated. The meteorological indicators that have a significant impact on the plum rain season were comprehensively considered. Based on the principle of membership degree in fuzzy mathematics, the qualitative evaluation of meteorological characteristic quantities that meet the requirements for the onset of the plum rain season was converted into a quantitative evaluation.

[0053] The specific process is as follows: Let T(i), determine the membership of each feature, and establish that the membership of meeting, approaching, and not meeting the conditions for the arrival of the plum rain season are 1, 0, and -1, respectively. The key feature N of the present invention is 9, and the extraction method and membership of each factor are shown in Table 2:

[0054] Table 2 Feature extraction method and membership assignment rules:

[0055] ;

[0056] Table 2 continued:

[0057] ;

[0058] In the table, Indicates 200hPa zonal wind; Indicates 500hPa geopotential height; represents the latent heat of water vapor; represents the specific heat capacity of air; Indicates absolute temperature; Indicates vertical speed; represents relative humidity; T represents temperature.

[0059] Step 2: Construct the credibility index of the plum rain forecast. The credibility index is obtained by weighted summation of the membership of each characteristic quantity, and the weight distribution is determined by the hierarchical analysis method.

[0060] Constructing a credibility index for the onset of plum rain season prediction based on circulation patterns and factors: , where: T represents the reliability of whether the conditions for the onset of plum rain are met (the closer T is to 7, the more favorable the onset of plum rain); T(G) represents the membership of the subtropical high ridge; T(K) represents the membership of the South Asian high ridge; T(t) represents the membership of -8℃ at 500hPa at 35N, 120E; T(r) represents the membership of 80% relative humidity in the (31-34, 110-120) area; T(h) represents the membership of the westerly jet center at 200hPa; T(l) represents the membership of the 850hPa low-level jet center in the range of 110-120E; T(w) represents the membership of the wind direction in Kolkata at 500hPa; 、 、 、 、 、 、 Represent the weight coefficients of each item respectively.

[0061] A judgment matrix (Table 3) was constructed based on the expert experience of each feature quantity, and the consistency ratio CR was calculated to conduct a consistency test.

[0062] Table 3 Constructing the judgment matrix of each feature quantity based on expert scoring

[0063] ;

[0064] Table 3 is used to calculate the consistency index CI = (λmax - n) / (n - 1); where n is the order of the judgment matrix and λmax is the maximum eigenvalue of the matrix.

[0065] Calculate the consistency ratio CR = CI / RI. RI values ​​are usually obtained by looking up tables, or by referring to relevant literature or materials.

[0066] If CR < 0.1, it is considered that the consistency of the judgment matrix is ​​acceptable and the weight distribution is reasonable.

[0067] If CR ≥ 0.1, the consistency of the judgment matrix is ​​considered to be poor and the judgment matrix needs to be modified.

[0068] On the basis of meeting the consistency test, the characteristic vector method is used to calculate the weight of each characteristic quantity and obtain 、 、 、 、 、 、 Weight coefficient.

[0069] The above calculation is based on the membership of each feature quantity to calculate the credibility index of the arrival of plum rain on each day, determine whether the plum rain category is on each day, and form a historical data set of the membership of each factor feature quantity, the credibility index, and the arrival of plum rain category. The training data set and the test data set are constructed with a ratio of 80% and 20% respectively. The data structure is as follows Figure 2 As shown in Figure 2, the historical dataset structure of key feature quantity membership, credibility index and classification labels is displayed.

[0070] Step 3: Based on the long-sequence historical data set, use support vector machine, decision tree or logistic regression algorithm to train the plum rain classification prediction model. The input of the model includes the credibility index and the key feature quantity membership. Figure 3 As shown in the figure, it includes input layer, machine learning algorithm layer (support vector machine, decision tree, etc.), model evaluation layer and prediction network layer.

[0071] S3.1 uses the actual data of the arrival of plum rain in Jianghuai from 1991 to 2020 (actual record data of plum rain from Jiangsu Provincial Meteorological Observatory), ERA5 analysis data (global meteorological reanalysis data set released by the European Centre for Medium-Range Weather Forecasts (ECMWF)) and the corresponding characteristic quantity algorithm to calculate and extract different characteristic quantities, and establish a long sequence of key characteristic quantity data of plum rain meteorological factors based on different characteristic quantity membership thresholds. Each year, 10 days before and after the arrival of plum rain are selected to calculate the daily plum rain prediction credibility index. According to the historical plum rain arrival date, the number before the arrival of plum rain is -1, and the number after the arrival of plum rain is 1. It is divided into two categories, forming a judgment case library with 20*71, and forming a long sequence data set containing 7 key characteristic quantity memberships, credibility indexes, and classifications of whether plum rain has arrived. Figure 4 As shown in the figure, the feature extraction process based on ERA5 reanalysis data and actual data is demonstrated.

[0072] S3.2 builds a prediction model for the start of the plum rain season based on support vector machines and decision trees. This method consists of four layers:

[0073] The first layer: input layer, input preprocessed training data set X and related classification Y.

[0074] The second layer: The data processing network layer of different machine learning methods trains the dataset X and the related classification Y through different methods. The steps of different methods are as follows:

[0075] Support Vector Machine (SVM): Choose Radial Basis Function (RBF) kernel or Polynomial kernel to increase data dimensionality and explore complex data relationships, such as Figure 5 As shown, the hyperplane partitioning and kernel function mapping process are demonstrated.

[0076] Its equation can be expressed as: ,in, is the hyperplane normal vector, b is the intercept, for any point x, if: , it is classified as the positive class; , it is classified as negative. The polynomial kernel function and the optimized parameters are used to train the SVM model and determine the support vector.

[0077] The polynomial kernel function formula is as follows: Where, and represents the input vector, represents the dot product of the two vectors, and d is the degree of the polynomial kernel, which needs to be selected according to the specific problem.

[0078] Decision tree: Select the improved CART and use the Gini index as the splitting criterion. For a dataset D containing K categories, the calculation formula for its Gini index is:

[0079] ;

[0080] in, Represents the proportion of category k in data set D, D represents the current data set, A represents a feature, and is used to split data set D into two subsets and , represents the total number of samples in the dataset D, and Represent subsets respectively and The number of samples in the Gini coefficient ( ) and the Gini coefficient ( ) represent subsets respectively and The Gini index ranges from [0, 1], and the smaller the value, the higher the purity of the dataset.

[0081] The third layer: model evaluation layer, which uses test set data to evaluate the trained prediction model. Common indicators include accuracy, recall, F1 score, etc.

[0082] , ,

[0083] , ;

[0084] TP represents the number of samples that the model correctly predicts as positive class;

[0085] TN represents the number of samples that the model correctly predicts as negative categories;

[0086] FP represents the number of samples that the model incorrectly predicts as positive.

[0087] FN represents the number of samples where the model incorrectly predicts the positive category as the negative category;

[0088] Precision indicates how many of the samples predicted by the model as positive samples are actually positive samples.

[0089] The fourth layer: prediction network layer, which determines the final prediction model of the two types of methods based on different methods and tests.

[0090] Step 4: Obtain the circulation element data of the global forecast system in real time, extract key feature quantities and calculate their membership, and input them into the prediction model to classify and judge the conditions for the onset of plum rain season.

[0091] S4.1 determines the spatial range for readings based on the calculation of different characteristic quantities. For example, relative humidity, low-level jet stream, and westerly jet stream are read at different latitudes within the 110-120°E range. The northeastern cold vortex is read at the average 500hPa altitude in the 40-50°N, 125-135°E region. The Kolkata wind is the wind direction at the station at 500hPa. A temperature reading of -8°C at 500hPa is a temperature change between 35°N and 120°E.

[0092] S4.2 uses algorithms for different characteristic quantities (subtropical high ridge, South Asian high ridge, westerly jet stream, pseudo-equivalent potential temperature) to intelligently identify and calculate and extract information of different characteristic quantities.

[0093] Method for determining the position of the 120E subtropical high ridgeline: the latitude of the subtropical high center within the range of 10-60N and the 120E meridian (within the 588 potential decameter line, zonal wind u=0, and du / dy>0).

[0094] Method for determining the position of the South Asian high pressure ridge: The average latitude of the positions at different longitudes within the range of 10-60N, 100-120E, and 1256 potential ten-meter lines, where the zonal wind u=0 and du / dy>0 are satisfied.

[0095] The method for calculating the position of the westerly jet stream is: the latitude corresponding to the maximum average value of the U wind values ​​at different grid points on the same latitude in the range of 20-60N in the 200hPa height field at 110-120E.

[0096] False equivalent potential temperature value: The 850hPa false equivalent potential temperature in the range of 20-60N is the average value of 110-120°E. When the 340K isovalue line is >30N, the Jianghuai plum rain conditions are met.

[0097] Kolkata Wind: Read the 500hPa actual wind direction and speed at the Kolkata station and the wind direction and speed at the nearest grid point of the forecast field.

[0098] S4.3 Use the hierarchical analysis method to comprehensively judge whether the circulation adjustment indicators and factor indicators can be met to meet the conditions for the beginning of the plum rain season, and objectively determine the circulation adjustment day.

[0099] The hierarchical analysis judgment criteria include: (1) The wind direction in Kolkata at 500hPa changes from westerly to easterly or alternating east and west winds. (2) The South Asian high pressure ridge line at 110-120E is greater than or equal to 28N, and the subtropical high pressure ridge line at 120E is greater than or equal to 20N. (3) The center of the 200hPa westerly jet stream jumps from 32-35N to 36-42N. (4) The centers of the 700 and 850hPa low-level jet streams in the range of 110-120E reach 28-30N, and the 80% humidity zone is located at 31-34N. (5) The 850hPa pseudo-equivalent potential temperature line in the range of 110-120E, 31-34N is greater than 340K. (6) The temperature at 35N and 120E at 500hPa is greater than -8℃. When all the above conditions are met, the circulation situation is adjusted and the conditions for the onset of plum rain are met. Specific analysis of the judgment order and the threshold values ​​of each indicator, such as Figure 6 As shown, the hierarchical judgment logic of key indicators such as the jump of the subtropical high ridge line and the northward movement of the westerly jet stream is demonstrated.

[0100] S5. Combine the actual monitoring and forecast data of ground elements to determine whether the plum rain monitoring standards are met, including the regional precipitation threshold, the proportion of rainy days and the average temperature conditions, and comprehensively determine the date of the start of the plum rain season.

[0101] On the basis of meeting the conditions for the onset of the plum rain season, with the circulation adjustment day as a necessary condition and ground factors as sufficient conditions, a comprehensive assessment of the ground factor conditions is carried out according to the plum rain monitoring standards, combined with the current precipitation situation and the model precipitation forecast for the next week. Specific criteria: (1) A rainy day is defined as one where more than one-third of the monitoring stations in the monitoring area experience precipitation greater than or equal to 0.1 mm, and the average daily precipitation in the area is greater than or equal to 2.0 mm. (2) Starting from the first rainy day, the proportion of moderate rainy days in the next 10 days accounts for greater than or equal to 50% of the total number of days in the corresponding period. (3) The average daily temperature is greater than or equal to 22°C.

[0102] The calculation formula for the monitoring standard in the station area (south of the Huaihe River, between the Yangtze River and the Huaihe River, and southern Jiangsu) is:

[0103] Temperature conditions: ;

[0104] Regional precipitation conditions: ;

[0105] Conditions for persistence of rainy days: , Meet the rainy day conditions;

[0106] Where, Indicates the average temperature of the monitoring stations within the region; represents the average precipitation at the stations within the region; represents the total number of stations in the monitoring area; N represents the number of stations in the region where the precipitation at a single station is greater than or equal to 0.1 mm; Indicates the total precipitation amount for the next 10 days.

[0107] Example: An intelligent identification method for determining the onset of the plum rain season in different regions of the Jianghuai region is formulated based on the monitoring standards for the onset of the plum rain season. Using real-time monitoring and model precipitation forecasts, an item-by-item judgment method is adopted to achieve intelligent judgment of the onset of the plum rain season.

[0108] The specific steps are as follows:

[0109] a: Based on the monitoring stations and standards for plum rain in Jiangsu, the monitoring standards for the onset of plum rain were standardized and objectively defined in steps to lay the foundation for intelligent identification. Representative stations were identified in the south of the Huaihe River, between the Yangtze River and the Huaihe River, and in the middle and lower reaches of the Yangtze River, and the values ​​for one-third of these stations were set. The statistical elements met the following requirements: (1) more than one-third of the monitoring stations in the monitoring area experienced precipitation greater than or equal to 0.1 mm, and the average daily precipitation in the area was greater than or equal to 2 mm; (2) from the first rainy day, the proportion of moderate rainy days in the next 10 days was greater than or equal to 50%. (3) The average temperature was greater than or equal to 22°C.

[0110] b: Interpolate the gridded precipitation forecast and temperature forecast to the Meiyu monitoring station to achieve a perfect fusion of real-time monitoring and forecast fields, and form time series data of precipitation and temperature.

[0111] c: Construct an automatic algorithm for each factor to determine whether the conditions for the beginning of the plum rain season are met every day. If the three conditions in a are met, the first rainy day will be determined as the beginning of the plum rain season under the monitoring standard.

[0112] When the circulation adjustment day obtained separately occurs before the beginning of the plum rain season under the monitoring standard, the beginning of the plum rain season under the monitoring standard is the official beginning of the plum rain season in the area. When the two occur on the same date, the circulation adjustment day is the beginning of the plum rain season in the area.

[0113] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiment. All technical solutions based on the concept of the present invention are within the scope of protection of the present invention. It should be noted that for those skilled in the art, various improvements and modifications that do not depart from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. An intelligent forecasting method for the onset of plum rain in a region based on circulation adjustment and factor judgment is characterized by: The method comprises the following steps: S1. Determine multiple key characteristic quantities that affect the onset of plum rain, and calculate the membership degree of each characteristic quantity based on historical data; S2. Constructing a credibility index for the prediction of the onset of the plum rain season. The credibility index is obtained by weighted summation of the membership of each characteristic quantity, and the weight distribution is determined by the hierarchical analysis method; S3. Based on the historical data set from 1991 to 2020, a support vector machine, decision tree, or logistic regression algorithm is used to train a plum rain classification prediction model, where the input of the model includes the credibility index and the key feature quantity membership; S4. Real-time acquisition of circulation element data from the global forecast system, extraction of key characteristic quantities, calculation of their membership and credibility index, and input into the forecast model for classification and judgment of plum rain season conditions; S5. Combine the actual monitoring and forecast data of ground elements to determine whether the plum rain monitoring standards are met, including the regional precipitation threshold, the proportion of rainy days and the average temperature conditions, and comprehensively determine the date of the start of the plum rain season.

2. The method for intelligently forecasting the onset of the plum rain season in a region based on circulation adjustment and factor judgment according to claim 1 is characterized in that: Several key characteristic quantities that affect the onset of plum rain include: 500hPa Kolkata wind direction, subtropical high pressure ridge, South Asian high pressure ridge, westerly jet stream, northeastern cold vortex, 850hPa pseudo-equivalent potential temperature, 850hPa low-level jet stream, 850hPa humidity, and 500hPa-8℃ line.

3. The method for intelligently forecasting the onset of the plum rain season in a region based on circulation adjustment and factor judgment according to claim 2 is characterized in that: The method for determining the position of the subtropical high ridge line includes: identifying the subtropical high center within the 588 geopotential decadal line within the range of 10-60°N and 120°E meridian, and determining the latitude value when its zonal wind u=0 and du / dy>0; The method for determining the position of the South Asian high pressure ridge includes: calculating the average of the latitude values ​​at different longitudes within the range of 1256 potential decameter lines in the range of 10-60°N and 100-120°E, which satisfies the zonal wind u=0 and du / dy>0.

4. The method for intelligently forecasting the onset of the plum rain season in a region based on circulation adjustment and factor judgment according to claim 1 is characterized in that: The assignment rule of membership is: the membership of the feature quantity that meets the conditions for the arrival of the plum rain season is 1, the membership of the feature quantity that is close to the conditions for the arrival of the plum rain season is 0, and the membership of the feature quantity that does not meet the conditions for the arrival of the plum rain season is -1.

5. The method for intelligently forecasting the onset of the plum rain season in a region based on circulation adjustment and factor judgment according to claim 1 is characterized in that: The process of weight allocation using the hierarchical analysis method is as follows: construct a judgment matrix through expert experience, calculate the weight value of each feature value, and perform a consistency test. When the consistency ratio is less than 0.1, the weight allocation is considered valid.

6. The method for intelligently forecasting the onset of the plum rain season in a region based on circulation adjustment and factor judgment according to claim 1 is characterized in that: The support vector machine uses a radial basis function kernel or a polynomial kernel, and the formula of the polynomial kernel is: ;in, and represents the input vector, represents the dot product of these two vectors, and d is the degree of the polynomial kernel.

7. The method for intelligently forecasting the onset of the plum rain season in a region based on circulation adjustment and factor judgment according to claim 1 is characterized in that: The decision tree uses the CART algorithm and uses the Gini index as the splitting criterion. The calculation formula of the Gini index is: ； in, Represents the proportion of category k in data set D, D represents the current data set, A represents a feature, and is used to split data set D into two subsets and , represents the total number of samples in the dataset D, and Respectively represent subsets and The number of samples in the Gini coefficient ( ) and the Gini coefficient ( ) represent subsets respectively and The Gini coefficient of the Gini index ranges from [0,1]. The smaller the value, the higher the purity of the data set.

8. The method for intelligently forecasting the onset of the plum rain season in a region based on circulation adjustment and factor judgment according to claim 1 is characterized in that: Real-time circulation element data, including ERA5 reanalysis data, are obtained, and forecast data are matched to monitoring sites based on the grid interpolation method.

9. The method for intelligently forecasting the onset of the plum rain season in a region based on circulation adjustment and factor judgment according to claim 1 is characterized in that: Meiyu monitoring standards further include: a. More than one-third of the stations in the monitoring area have daily precipitation ≥ 0.1 mm, and the regional average daily precipitation is ≥ 2.0 mm; b. The proportion of rainy days within 10 days from the first rainy day is ≥50%; c. Daily average temperature ≥ 22℃.

10. The method for intelligently forecasting the onset of the plum rain season in a region based on circulation adjustment and factor judgment according to claim 1 is characterized in that: The rule for comprehensively determining the date of the onset of the plum rain season is as follows: the circulation adjustment date is the date when the circulation conditions for the onset of the plum rain season are met through the analysis of the key characteristic quantities of the circulation elements in step S4; if the circulation adjustment date is earlier than the date when the ground element conditions are met in step S5, the date when the ground element conditions are met is used as the date of the onset of the plum rain season; If the two dates are the same, the circulation adjustment day is used as the beginning of the plum rain season. The determination of the circulation adjustment day is combined with the classification judgment result of the plum rain season condition of the prediction model in step S4.

Images