Rainstorm early warning threshold calculation method and device, computer device and storage medium

Abstract

The application relates to the field of finance and insurance, and discloses a calculation method and device of a rainstorm early warning threshold, computer equipment and a storage medium. The method comprises the following steps: acquiring rainfall data of a target area in a preset time period and claim data caused by a rainstorm; calculating a rainstorm risk probability of a rainstorm event and an occurrence event according to the rainfall data and the claim data when an estimated rainstorm early warning threshold is different, wherein the rainstorm event is an event in which actual rainfall exceeds the corresponding estimated rainstorm early warning threshold; and determining the estimated rainstorm early warning threshold corresponding to the maximum rainstorm risk probability as the final rainstorm early warning threshold of the target area. The application optimizes the calculation method of the rainstorm early warning threshold, so that the rainstorm early warning threshold is selected accurately and has regional pertinence.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of financial insurance data processing, and in particular to a calculation method and device of a rainstorm early warning threshold, a computer device and a storage medium. BACKGROUND

[0002] In the financial insurance industry, there are many insurance types related to rainstorms, such as property all-risk insurance and agricultural insurance clauses containing rainstorm liability. For insurance companies, timely early warning of customers before a rainstorm can reduce losses for both parties.

[0003] In the prior art, the existing rainstorm early warning threshold in the financial industry such as insurance mostly adopts a national unified standard threshold, without considering the differences between regions or the actual situation, and lacks pertinence. Moreover, since the amount of rainfall has no direct causal relationship with whether or not to be at risk, selecting a fixed threshold to trigger rainstorm early warning by the national rainfall standard will lead to problems of untimely early warning or excessive early warning in different regions. SUMMARY

[0004] To solve the technical problem of untimely early warning or excessive early warning caused by using a unified fixed threshold for early warning in the prior art, the present application provides a calculation method and device of a rainstorm early warning threshold, a computer device and a storage medium, which mainly aims to optimize the calculation method of the rainstorm early warning threshold, so that the selection of the rainstorm early warning threshold is rigorous, accurate and regionally pertinent.

[0005] To achieve the above-mentioned purpose, the present application provides a calculation method of a rainstorm early warning threshold, which comprises:

[0006] obtaining rainfall data of a target region in a preset time period and claim data caused by rainstorm leading to risk;

[0007] calculating, according to the rainfall data and the claim data, rainstorm risk probabilities of the rainstorm events and the risk events existing when the estimated rainstorm early warning threshold is different values, wherein the rainstorm event is an event in which the actual rainfall exceeds the corresponding estimated rainstorm early warning threshold;

[0008] determining the estimated rainstorm early warning threshold corresponding to the maximum rainstorm risk probability as the final rainstorm early warning threshold of the target region.

[0009] In addition, to achieve the above-mentioned purpose, the present application also provides a calculation device of a rainstorm early warning threshold, which comprises:

[0010] a data acquisition module for obtaining rainfall data of a target region in a preset time period and claim data caused by rainstorm leading to risk;

[0011] The computing module is configured to calculate a rainstorm risk probability of a rainstorm event and an occurrence of a risk event according to the rainfall data and the claim data, wherein the rainstorm event refers to an event in which the actual rainfall exceeds the corresponding estimated rainstorm early warning threshold.

[0012] The threshold determination module is configured to determine the estimated rainstorm early warning threshold corresponding to the maximum rainstorm risk probability as the final rainstorm early warning threshold of the target region.

[0013] To achieve the above object, the present application further provides a computer device comprising a memory, a processor, and computer readable instructions stored in the memory and executable on the processor, wherein the processor executes the steps of the rainstorm early warning threshold calculation method according to any one of the preceding embodiments.

[0014] To achieve the above object, the present application further provides a computer readable storage medium having computer readable instructions stored thereon, wherein the computer readable instructions are executable on a processor to cause the processor to execute the steps of the rainstorm early warning threshold calculation method according to any one of the preceding embodiments.

[0015] The rainstorm early warning threshold calculation method, device, computer device, and storage medium provided by the present application optimize the existing rainstorm early warning threshold selection, calculate the final rainstorm early warning threshold of the target region according to the historical rainfall data and the historical claim data, make the early warning standard reasonably set in combination with the local geographical characteristics, and avoid the occurrence of early warning delay or early warning overkill in different regions caused by a single threshold. The optimal rainstorm early warning threshold of the present embodiment is simple to implement, logically rigorous, can calculate the optimal rainstorm early warning threshold of any region, and has a wide range of applications. With the update of the rainfall data and the claim data of the target region, the final rainstorm early warning threshold can be updated adaptively, and the environment can be flexibly adapted. The accurate calculation of the final rainstorm early warning threshold by the present embodiment can timely warn the customers, reduce the loss of the customers and the claim risk of the insurance company, and achieve a win-win situation. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 FIG. 1 is a flowchart of the rainstorm early warning threshold calculation method according to an embodiment of the present application;

[0017] Figure 2 FIG. 2 is a structural block diagram of the rainstorm early warning threshold calculation device according to an embodiment of the present application;

[0018] Figure 3 FIG. 3 is an internal structural block diagram of the computer device according to an embodiment of the present application.

[0019] The implementation, functional features, and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0020] The technical solutions and advantages of the embodiments of the present application will be more clearly understood from the following description of the embodiments of the present application with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application. It should be understood that the specific embodiments described herein are only used to explain the present application and should not be used to limit the present application.

[0021] Figure 1 FIG. 1 is a flowchart of a method for calculating a heavy rain warning threshold according to an embodiment of the present application. As shown in FIG. 1, the method for calculating the heavy rain warning threshold includes the following steps S100-S300. Figure 1

[0022] S100: Obtain rainfall data of a target area in a preset time period and claim data caused by heavy rain.

[0023] Specifically, the target area can be a geographical division of a country (for example, East China, South China, North China, Central China, Southwest China, Northwest China and Northeast China), a province (for example, Guangdong Province, Jiangsu Province, etc.), a city (for example, Beijing, Shanghai, Shenzhen, Guangzhou, etc.), an administrative block of a city (for example, Beijing includes multiple districts such as Chaoyang District, Haidian District, Dongcheng District, Xicheng District, etc.), and the like, but is not limited thereto.

[0024] The preset time period is a time range, which is configured according to actual conditions, for example, 2012-2021, a total of 10 years, and the present application does not limit this.

[0025] The rainfall data and the claim data are both historical data. The rainfall data includes the rainfall of each sub-region of the target area in each day of the preset time period. If the target area is large, the rainfall of different sub-regions of the target area in the same day may not be consistent and may have large differences. For example, if the target area is a province, the rainfall of different cities in the same day of the province is different. In order to reduce the interference of regional differences, the target area is divided into multiple sub-regions.

[0026] ​More specifically, the rainfall data can be the European Centre for Medium-Range Weather Forecasts (ECMWF) Fifth Generation Atmospheric Reanalysis Data (ERA5) grid rainfall data, which is a comprehensive reanalysis data. The grid (grid) of the embodiment can be divided at a spatial resolution of 10 kilometers (0.1 degree), that is, divided into a 0.1 degree x 0.1 degree grid, but is not limited thereto. The time resolution is daily time interval, and the time range is from 2012 to 2021. One sub-region corresponds to one grid.

[0027] wherein ERA5 is the fifth generation atmospheric reanalysis meteorological data of ECMWF for global climate, and the format is generally Grid and NetCDF format, and the data in Grib format and NetCDF format can be converted into common data format.

[0028] The ERA5 data is meteorological data, supporting real-time data from 1979 to now, with time resolution of hour level, day level, month level, etc., spatial resolution of about 0.25°-0.5°, and containing common meteorological data such as humidity, wind speed, temperature, and precipitation. The reanalysis combines model data with observations from around the world to form a global complete and consistent dataset. ERA5 replaces its predecessor, the ERA-Interim reanalysis.

[0029] The claim data is historical claim data caused by heavy rain in the target area within the same preset time period. The claim data is the data generated by the insurance compensation event (claim event) of the insurance company for the insured user.

[0030] S200: According to the rainfall data and the claim data, the heavy rain claim probability of a heavy rain event and an existing claim event is calculated when the estimated heavy rain warning threshold is different, wherein the heavy rain event is an event in which the actual rainfall exceeds the corresponding estimated heavy rain warning threshold.

[0031] Specifically, the estimated heavy rain warning threshold is a pre-set heavy rain warning threshold. The embodiment sets a plurality of different estimated heavy rain warning thresholds, sets the claim event as A, sets the heavy rain event as B, and sets the estimated heavy rain warning threshold as C. The value of C can be, for example, different values in the array [20, 30, 40, …, 150], wherein each value of C is in millimeters (mm). The heavy rain event B is an event in which the actual daily rainfall exceeds the estimated heavy rain warning threshold, or an event in which the daily rainfall of continuous multiple days exceeds the estimated heavy rain warning threshold.

[0032] The estimated rainstorm early warning threshold directly affects the definition of a rainstorm event. Different setting of the estimated rainstorm early warning threshold setting C can define different days or times of rainstorm events. For example, if C = 50 mm, a rainstorm event occurs on a certain day when the daily rainfall is 60 mm; if C = 70 mm, a rainstorm event does not occur on a certain day when the daily rainfall is 60 mm.

[0033] S300: Determine the estimated rainstorm early warning threshold corresponding to the maximum rainstorm risk probability as the final rainstorm early warning threshold of the target region.

[0034] Specifically, the embodiment designs a new method for evaluating regional rainstorm early warning thresholds. By calculating the risk probability of a certain region under different estimated rainstorm early warning thresholds, the estimated rainstorm early warning threshold corresponding to the maximum risk probability is found as the final rainstorm early warning threshold of the region.

[0035] The rainstorm risk probability is how likely a certain target will cause a claim due to rainstorm. When the rainstorm risk probability is maximum, it means that the correlation between rainstorm and claim is strongest. Finding the maximum rainstorm early warning threshold means that it can be inferred when an accident is most likely to occur. Therefore, once the rainfall in the region reaches this maximum rainstorm early warning threshold, an early warning can be sent to the corresponding customer to minimize losses. The larger the threshold, the higher the early warning standard, and it is not easy to issue an early warning. The smaller the threshold, the lower the early warning standard, and it is easier to issue an early warning. The final rainstorm early warning threshold of the embodiment is based on the existing customer claim data to summarize experience and determine how much rainfall to warn customers after it rains. The final rainstorm early warning threshold is reasonably set and will not be too large or too small, effectively balancing excessive early warning and delayed early warning, and ensuring timely and effective early warning.

[0036] The embodiment optimizes the selection of the existing rainstorm early warning threshold. The final rainstorm early warning threshold of the target region is calculated based on historical rainfall data and historical claim data, so that the early warning standard is reasonably set in combination with local geographical characteristics, avoiding the occurrence of early warning delay or excessive early warning in different regions caused by a single threshold. The best rainstorm early warning threshold of the embodiment is simple to implement and has rigorous logic, can calculate the best rainstorm early warning threshold of any region, and has a wide range of applications. With the update of rainfall data and claim data of the target region, the final rainstorm early warning threshold can be updated adaptively, and thus flexibly adapt to changing environments. Precise calculation of the final rainstorm early warning threshold by the embodiment can timely warn customers, reduce customer losses and insurance company claim risks, and achieve a win-win situation.

[0037] In one embodiment, the rainfall data includes sub-rainfall data of each sub-region of the target region within a preset time period, and the claim data includes sub-claim data of each sub-region within the preset time period.

[0038] Step S200 specifically comprises:

[0039] According to the sub-rain data and the sub-claim data, the number of risky rainstorms of each sub-region occurring in the event and existing in the rainstorm event is calculated when the estimated rainstorm warning threshold value is the same value.

[0040] According to the sub-rain data, the number of rainstorm events of each sub-region occurring in the rainstorm event is calculated when the estimated rainstorm warning threshold value is the same value.

[0041] The sum of the number of risky rainstorms of all target sub-regions corresponding to the same value of the estimated rainstorm warning threshold value is summed up, and the sum of the number of rainstorm events of all target sub-regions is summed up, and the ratio of the sum of the number of risky rainstorms to the sum of the number of rainstorm events is calculated as the rainstorm risk probability corresponding to the estimated rainstorm warning threshold value.

[0042] Specifically, the target region includes a plurality of sub-regions, and each sub-region corresponds to a grid. The rainfall data includes sub-rain data corresponding to each grid. In this embodiment, the sub-rain data of each grid, i.e. each sub-region, in a preset time period is selected. This embodiment is preferably daily with a time resolution of a time interval. Each sub-rain data includes the rainfall data of the corresponding sub-region every day in the preset time period.

[0043] Similarly, each sub-region also has corresponding sub-claim data, and the sub-claim data includes the data of the corresponding sub-region every day in the same preset time period, such as the number of risky cases.

[0044] Different estimated rainstorm warning threshold values correspond to a rainstorm risk probability, and the rainstorm risk probability corresponds to the entire target region. Since the entire target region is relatively extensive, the rainfall conditions of different sub-regions are not completely the same at the same time, therefore, the sub-rain data and the sub-claim data of the sub-regions are aggregated to obtain the rainstorm risk probability of the entire target region.

[0045] The estimated rainstorm warning threshold value C has a plurality of different values, and this embodiment calculates the number of risky rainstorms of each sub-region occurring in the event B when the event A occurs under different values of C, denoted as N(B|A). This embodiment also calculates the number of rainstorm events of each sub-region occurring in the event B under different values of C, denoted as N(B). Since the values of C are different, N(B) may not be equal under different values of C.

[0046] The sum of the risk rain times N(B|A) of all target sub-regions is obtained by summing the risk rain times N(B|A) of all target sub-regions under the same value of the estimated rainstorm early warning threshold C. The sum of the rainstorm event times N(B) of all target sub-regions is obtained by summing the rainstorm event times N(B) of all target sub-regions under the same value of the estimated rainstorm early warning threshold C. The ratio of the sum of the risk rain times to the sum of the rainstorm event times is taken as the rainstorm risk probability corresponding to the estimated rainstorm early warning threshold. The specific formula is as follows:

[0047]

[0048] Wherein, P(A|B) is the rainstorm risk probability corresponding to an estimated rainstorm early warning threshold, N(A|B) i is the risk rain time of the i-th target sub-region, N(B) i is the rainstorm event time of the i-th target sub-region, and i is 1, 2,..., L, and L is the number of target sub-regions. The target sub-region is part of the sub-regions or all sub-regions.

[0049] In this embodiment, the risk rain times and the rainstorm event times of all sub-regions under the same value of the estimated rainstorm early warning threshold are aggregated and counted, and the ratio of the two is taken as the rainstorm risk probability corresponding to the estimated rainstorm early warning threshold. The rainstorm risk probability is calculated by comprehensively considering the rainstorm claim situation of all sub-regions, which is simple, reasonable, accurate and logically rigorous.

[0050] In one embodiment, the rain data includes sub-rain data of each sub-region of the target region in a preset time period, and the claim data includes sub-claim data of each sub-region in the preset time period.

[0051] Step S200 specifically includes:

[0052] According to the sub-rain data and the sub-claim data, the sub-rainstorm risk probability of each sub-region under the same value of the estimated rainstorm early warning threshold is calculated.

[0053] The sub-rainstorm risk probability of all target sub-regions corresponding to the same value of the estimated rainstorm early warning threshold is averaged to obtain the rainstorm risk probability corresponding to the estimated rainstorm early warning threshold.

[0054] Specifically, the target region includes a plurality of sub-regions, and each sub-region corresponds to a grid. The rain data includes sub-rain data corresponding to each grid. In this embodiment, the sub-rain data of each grid, i.e., each sub-region in a preset time period is selected. In this embodiment, the time resolution is preferably a time interval per day. Each sub-rain data includes the rainfall data of the corresponding sub-region every day in the preset time period.

[0055] Similarly, each sub-region also has corresponding sub-claim data, and the sub-claim data includes the daily claim data, for example, the number of claims, of the corresponding sub-region in the same preset time period.

[0056] Different estimated heavy rain warning thresholds correspond to a heavy rain claim probability, and the heavy rain claim probability corresponds to the entire target region. Since the entire target region is relatively extensive, the rainfall conditions of different sub-regions are not completely the same at the same time. Therefore, the sub-rain data and the sub-claim data of the sub-regions are aggregated to obtain the heavy rain claim probability of the entire target region.

[0057] More specifically, the estimated heavy rain warning threshold C has multiple different values, and the sub-heavy rain claim probability of each sub-region to occur the heavy rain event B and the claim event A is calculated under different values of C, denoted as Q(A|B). Since the values of C are different, the Q(A|B) corresponding to the same sub-region under different values of C can be different.

[0058] The average of the sub-heavy rain claim probability Q(A|B) of all target sub-regions under the same value of the estimated heavy rain warning threshold C is calculated, and the obtained average is taken as the heavy rain claim probability P(A|B) corresponding to the estimated heavy rain warning threshold. The specific formula is as follows:

[0059]

[0060] Wherein, P(A|B) is the heavy rain claim probability corresponding to an estimated heavy rain warning threshold, Q(A|B) i is the sub-heavy rain claim probability of the i-th target sub-region, i takes the value of 1, 2,..., L, and L is the number of target sub-regions. The target sub-region is part of the sub-regions or all the sub-regions.

[0061] In this embodiment, the sub-heavy rain claim probability of all sub-regions under the same value of the estimated heavy rain warning threshold is aggregated, the heavy rain claim probability corresponding to the estimated heavy rain warning threshold is obtained by calculating the average, and the rainfall claim conditions of all sub-regions are comprehensively considered. The calculation is simple, reasonable, accurate and logically rigorous.

[0062] In one embodiment, according to the sub-rain data and the sub-claim data, the sub-heavy rain claim probability of each sub-region to occur the heavy rain event and the claim event when the estimated heavy rain warning threshold is the same value is calculated, including:

[0063] According to the sub-rain data and the sub-claim data, the out-of-risk rain probability of each sub-region to occur the claim event and the heavy rain event when the estimated heavy rain warning threshold is the same value is calculated, and the actual out-of-risk probability of the claim event and the heavy rain probability of the heavy rain event are calculated.

[0064] According to the rainstorm probability, the actual risk probability and the rainstorm probability, the sub-rainstorm risk probability of each sub-region in the case of occurrence of a rainstorm event and occurrence of a risk event is calculated when the estimated rainstorm warning threshold value is a same value.

[0065] Specifically, the estimated rainstorm warning threshold value C has multiple different values, and the sub-rainstorm risk probability Q(A|B) of a same sub-region under different values of C can be different.

[0066] In the case that the estimated rainstorm warning threshold value C is a same value, the rainstorm probability Q(B|A) of occurrence of a rainstorm event B in the case of occurrence of a risk event A of each sub-region, the actual risk probability Q(A) of occurrence of a risk event A of each sub-region and the rainstorm probability Q(B) of occurrence of a rainstorm event B are calculated.

[0067] According to the rainstorm probability Q(B|A), the actual risk probability Q(A) and the rainstorm probability Q(A) of a same sub-region when the estimated rainstorm warning threshold value C is a same value, the sub-rainstorm risk probability Q(A|B) of occurrence of a risk event A in the case of occurrence of a rainstorm event B of the sub-region under the estimated rainstorm warning threshold value is calculated.

[0068] The specific formula is as follows:

[0069]

[0070] Wherein, Q(A|B) is the sub-rainstorm risk probability of occurrence of a risk event A in the case of occurrence of a rainstorm event B of a sub-region, Q(A) is the actual risk probability of occurrence of a risk event A of a sub-region, and Q(B) is the rainstorm probability of occurrence of a rainstorm event B of a sub-region. The sub-rainstorm risk probability Q(A|B) of any sub-target region i can be calculated by using formula (3). i

[0071] In one embodiment, according to the sub-rainfall data and the sub-claim data, the rainstorm probability of occurrence of a rainstorm event in the case of occurrence of a risk event of each sub-region when the estimated rainstorm warning threshold value is a same value is calculated, and the actual risk probability of occurrence of a risk event and the rainstorm probability of occurrence of a rainstorm event are calculated, including:

[0072] According to the sub-rainfall data, the rainfall times of each sub-region in a preset time period and the rainstorm times of occurrence of a rainstorm event when the estimated rainstorm warning threshold value is a same value are calculated.

[0073] According to the sub-claim data, the risk times of occurrence of a risk event of each sub-region are calculated.

[0074] ​According to the sub-claim data and the sub-rainfall data, the estimated heavy rain warning threshold is calculated as a same value, and the number of heavy rain events of each sub-region occurring in the risk event and the number of heavy rain events of each sub-region occurring in the risk event are calculated.

[0075] According to the number of heavy rain events of each sub-region occurring in the risk event and the number of heavy rain events of each sub-region occurring in the risk event, the number of heavy rain events of each sub-region occurring in the risk event is calculated under the estimated heavy rain warning threshold.

[0076] According to the number of heavy rain events of each sub-region occurring in the risk event and the number of heavy rain events of each sub-region occurring in the risk event, the number of heavy rain events of each sub-region occurring in the risk event is calculated under the estimated heavy rain warning threshold.

[0077] According to the number of heavy rain events of each sub-region occurring in the risk event and the number of heavy rain events of each sub-region occurring in the risk event, the number of heavy rain events of each sub-region occurring in the risk event is calculated under the estimated heavy rain warning threshold.

[0078] Specifically, the size of the estimated heavy rain warning threshold determines the determination of the heavy rain event. The same rainfall day of the same sub-region may be determined as a heavy rain event under some estimated heavy rain warning thresholds, and as a non-heavy rain event under other estimated heavy rain warning thresholds.

[0079] This embodiment will calculate the number of rainfall events R of each sub-region occurring in the risk event R in a preset time period, and the number of heavy rain events B of each sub-region occurring in the risk event B. Among them, the rainfall event R is irrelevant to the estimated heavy rain warning threshold C, and the heavy rain event B is related to the estimated heavy rain warning threshold C.

[0080] According to the sub-claim data of the sub-region, the number of risk events N(A) of the sub-region occurring in the risk event A can be calculated, and the number of risk events N(A) is irrelevant to the estimated heavy rain warning threshold C.

[0081] According to the sub-claim data and the sub-rainfall data of the sub-region, the number of heavy rain events N(B|A) of the sub-region occurring in the risk event A under each estimated heavy rain warning threshold is calculated.

[0082] When the estimated heavy rain warning threshold is a same value, according to the number of heavy rain events N(B|A) of the same sub-region and the number of risk events N(A), the risk rain probability Q(B|A) of the same sub-region under the estimated heavy rain warning threshold is calculated. The specific formula is as follows:

[0083]

[0084] According to the number of risk events N(A) and the number of rainfall events N(R) of the same sub-region, the actual risk probability Q(A) of the same sub-region is calculated. The specific formula is as follows:

[0085]

[0086] At the same value of the estimated rainstorm warning threshold, the rainstorm probability Q(B) of the sub-region under the estimated rainstorm warning threshold is calculated according to the rainstorm times N(B) and the rainfall times N(R) of the same sub-region. The specific formula is as follows:

[0087]

[0088] According to the formula (2) - formula (6),

[0089]

[0090] Therefore,

[0091]

[0092] Wherein, N(B|A)1 is the rainstorm times of the first target sub-region, N(B)1 is the rainstorm times of the first target sub-region, N(B|A)2 is the rainstorm times of the second target sub-region, N(B)2 is the rainstorm times of the second target sub-region, N(B|A) L is the rainstorm times of the Lth target sub-region, N(B) L is the rainstorm times of the Lth target sub-region.

[0093] The formula (1) - formula (8) is a direct application of the conditional probability in the Bayes formula.

[0094] The embodiment estimates the rainstorm probability, the actual risk probability and the rainstorm probability of the sub-region by the rainstorm times, the risk times, the rainfall times and the rainstorm times of the sub-region, which is simple, convenient and feasible.

[0095] In one embodiment, the rainfall data includes the actual daily rainfall of each day within a preset time period;

[0096] The rainstorm event is specifically that the event of the actual daily rainfall exceeding the corresponding estimated rainstorm warning threshold is recorded as a rainstorm event,

[0097] Or,

[0098] If there is no continuous multi-day actual daily rainfall exceeding the corresponding estimated rainstorm warning threshold, the event of the single-day actual daily rainfall exceeding the corresponding estimated rainstorm warning threshold is recorded as a rainstorm event,

[0099] If the continuous multi-day actual daily rainfall exceeds the corresponding estimated rainstorm warning threshold, the event of the continuous multi-day actual daily rainfall exceeding the corresponding estimated rainstorm warning threshold is recorded as a rainstorm event.

[0100] Specifically, considering that there is a certain delay and uncertainty in the record time of claims in continuous rainstorm events, errors may occur if statistics are calculated separately for each day, for example, the number of rainstorm events calculated for a single day is greater than the number of risk events calculated for the same time range.

[0101] Therefore, the present embodiment records a rainstorm event occurring in two or more consecutive days as one rainstorm event, and records a risk event occurring within the range as one risk event, so that the rainstorm event and the risk event are consistent in time statistics. For example, a rainstorm event occurs for a total of 3 days from 1.1 to 1.3 days, and the continuous rainstorm event from 1.1 to 1.3 days is recorded as one rainstorm event. The risk event from 1.1 to 1.3 days is also recorded as one risk event.

[0102] In addition, if a risk event occurs on a certain day, regardless of the number of risk events (customer reports, insurance company compensation), the corresponding event on that day is recorded as one risk event, regardless of the number of risk events actually occurring on the same day. If no risk event occurs on a certain day, the corresponding event on that day is recorded as a non-risk event.

[0103] In one embodiment,

[0104] The target sub-region is all sub-regions contained in the target region,

[0105] Or,

[0106] The target sub-region is the sub-region remaining after the interference region is removed, wherein the interference region is a sub-region determined according to the sub-claim data, in which the number of claims in a preset time period is less than a minimum claim threshold, or the interference region is a sub-region determined according to the obtained insurance data, in which the number of insurance in a preset time period is less than a minimum insurance threshold.

[0107] Specifically, the target sub-region can include all sub-regions of the target region.

[0108] The target sub-region can also include part of the sub-regions of the target region. If the target region is very large, some sub-regions may be sparsely populated mountainous, rainforest, lake, river, etc. These regions may not have insurance or have few insured customers, and are non-insured regions; or these regions may have few claims, and are non-risk regions; these regions can be determined as interference regions. In this way, the calculation amount can be reduced, and the interference of accidental events can be excluded.

[0109] Of course, in actual application, the interference region can also be defined by other situations, which are not limited in the present application.

[0110] The application considers the actual insurance industry risk situation, and designs a new method for evaluating regional rainstorm early warning thresholds based on Bayesian probability for the insurance industry. The method is simple to implement and has rigorous logic. By calculating the rainstorm risk probability of a region (such as a province) under different estimated thresholds, the estimated threshold corresponding to the maximum rainstorm risk probability is found as the final rainstorm early warning threshold of the region, so that the business side of different regions corresponds to specific rainstorm early warning thresholds, and the rainstorm early warning is performed according to the corresponding rainstorm early warning threshold, effectively preventing the problems of untimely or excessive early warning, ensuring timely and effective early warning, reducing customer disaster losses, reducing the risk of risk, and realizing the win-win of customers and insurance companies.

[0111] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, and the execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the application.

[0112] Figure 3 The structure block diagram of the rainstorm early warning threshold calculation device in an embodiment of the application is shown in FIG. 1. Referring to FIG. 1, Figure 3 , the device comprises:

[0113] The data acquisition module 100 is configured to acquire rainfall data of a target region in a preset time period and claim data caused by rainstorm risk;

[0114] The calculation module 200 is configured to calculate, according to the rainfall data and the claim data, a rainstorm risk probability of a rainstorm event and an occurrence of a risk event when an estimated rainstorm early warning threshold is a different value, wherein the rainstorm event is an event that an actual rainfall exceeds the corresponding estimated rainstorm early warning threshold;

[0115] The threshold determination module 300 is configured to determine the estimated rainstorm early warning threshold corresponding to the maximum rainstorm risk probability as the final rainstorm early warning threshold of the target region.

[0116] In one embodiment, the rainfall data comprises sub-rainfall data of each sub-region of the target region in the preset time period, and the claim data comprises sub-claim data of each sub-region in the preset time period;

[0117] The calculation module 200 specifically comprises:

[0118] The first calculation module is configured to calculate, according to the sub-rainfall data and the sub-claim data, a risk rainstorm number of each sub-region when the estimated rainstorm early warning threshold is the same value and a risk event occurs and a rainstorm event exists;

[0119] The second calculation module is configured to calculate, according to the sub-rainfall data, a rainstorm event number of each sub-region when the estimated rainstorm early warning threshold is the same value and a rainstorm event occurs;

[0120] The third calculation module is configured to sum the numbers of rainstorm accidents of all target sub-regions corresponding to the same value of the estimated rainstorm early warning threshold, sum the numbers of rainstorm events of all target sub-regions, and calculate a ratio of the sum of the numbers of rainstorm accidents to the sum of the numbers of rainstorm events as a rainstorm accident probability corresponding to the estimated rainstorm early warning threshold.

[0121] In one embodiment, the rain data includes sub-rain data of each sub-region of the target region in a preset time period, and the claim data includes sub-claim data of each sub-region in the preset time period.

[0122] The calculation module 200 specifically includes:

[0123] The fourth calculation module is configured to calculate, according to the sub-rain data and the sub-claim data, a sub-rainstorm accident probability of each sub-region corresponding to the same value of the estimated rainstorm early warning threshold, in which a rainstorm event occurs and a rainstorm accident exists.

[0124] The fifth calculation module is configured to average the sub-rainstorm accident probabilities of all target sub-regions corresponding to the same value of the estimated rainstorm early warning threshold, to obtain the rainstorm accident probability corresponding to the estimated rainstorm early warning threshold.

[0125] In one embodiment, the fourth calculation module specifically includes:

[0126] The sixth calculation module is configured to calculate, according to the sub-rain data and the sub-claim data, a rainstorm accident probability of each sub-region corresponding to the same value of the estimated rainstorm early warning threshold, in which a rainstorm accident exists and a rainstorm event occurs, and calculate an actual rainstorm accident probability and a rainstorm probability of a rainstorm event.

[0127] The seventh calculation module is configured to calculate, according to the rainstorm accident probability, the actual rainstorm accident probability and the rainstorm probability, the sub-rainstorm accident probability of each sub-region corresponding to the same value of the estimated rainstorm early warning threshold, in which a rainstorm event occurs and a rainstorm accident exists.

[0128] In one embodiment, the sixth calculation module specifically includes:

[0129] The first sub-calculation module is configured to calculate, according to the sub-rain data, a rain number of each sub-region corresponding to the same value of the estimated rainstorm early warning threshold, in which a rain event occurs in a preset time period, and a rainstorm number of each sub-region corresponding to the same value of the estimated rainstorm early warning threshold, in which a rainstorm event occurs.

[0130] The second sub-calculation module is configured to calculate, according to the sub-claim data, a rainstorm accident number of each sub-region.

[0131] The third sub-calculation module is configured to calculate, according to the sub-rain data and the sub-claim data, a rainstorm accident number of each sub-region corresponding to the same value of the estimated rainstorm early warning threshold, in which a rainstorm accident exists and a rainstorm event occurs.

[0132] the fourth sub-computing module is configured to compute, according to the number of stormy weather events and the number of events in the same sub-region when the estimated storm warning threshold value is the same, a stormy weather probability of the sub-region under the estimated storm warning threshold value;

[0133] the fifth sub-computing module is configured to compute, according to the number of events and the number of rainfalls in the sub-region, an actual event probability of the sub-region;

[0134] the sixth sub-computing module is configured to compute, according to the number of stormy weather events and the number of rainfalls in the sub-region when the estimated storm warning threshold value is the same, a stormy weather probability of the sub-region under the estimated storm warning threshold value.

[0135] In one embodiment, the rainfall data includes actual daily rainfall in each day within a preset time period;

[0136] a stormy weather event is specifically an event in which actual daily rainfall exceeds a corresponding estimated storm warning threshold value,

[0137] or,

[0138] if there is no actual daily rainfall exceeding a corresponding estimated storm warning threshold value for consecutive days, an event in which actual daily rainfall exceeds a corresponding estimated storm warning threshold value on a single day is recorded as a stormy weather event,

[0139] if actual daily rainfall exceeds a corresponding estimated storm warning threshold value for consecutive days, an event in which actual daily rainfall exceeds a corresponding estimated storm warning threshold value for consecutive days is recorded as a stormy weather event.

[0140] In one embodiment, the target sub-region is all sub-regions included in the target region,

[0141] or,

[0142] the target sub-region is a sub-region remaining after interference regions are removed, wherein the interference regions are sub-regions determined according to sub-claim data and in which the number of claims is less than a minimum claim threshold value within a preset time period, or the interference regions are sub-regions determined according to obtained insurance data and in which the number of insurances is less than a minimum insurance threshold value within a preset time period.

[0143] The terms "first" and "second" in the above-mentioned modules / units are only used to distinguish different modules / units and are not intended to specify which module / unit has a higher priority or any other limiting meaning. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or modules is not necessarily limited to those steps or modules explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices. The module divisions appearing in this application are merely logical divisions; in actual applications, different division methods may be used.

[0144] Specific limitations regarding the calculation device for the rainstorm warning threshold can be found in the above description of the calculation method for the rainstorm warning threshold, and will not be repeated here. Each module in the aforementioned rainstorm warning threshold calculation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0145] Figure 3 This is a block diagram of the internal structure of a computer device according to an embodiment of this application. Figure 3 As shown, the computer device includes a processor, memory, network interface, input device, and display screen connected via a system bus. The processor provides computational and control capabilities. The memory includes storage media and internal memory. The storage media can be non-volatile or volatile. The storage media stores an operating system and may also store computer-readable instructions. When executed by the processor, these instructions enable the processor to calculate a rainstorm warning threshold. The internal memory provides an environment for the operation of the operating system and computer-readable instructions stored in the storage media. The internal memory may also store computer-readable instructions, which, when executed by the processor, enable the processor to perform the rainstorm warning threshold calculation. The network interface of the computer device is used for communication with an external server via a network connection. The display screen can be an LCD screen or an e-ink screen. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0146] In one embodiment, a computer device is provided, including a memory, a processor, and computer-readable instructions (e.g., a computer program) stored in the memory and executable on the processor. When the processor executes the computer-readable instructions, it implements the steps of the method for calculating the rainstorm warning threshold described in the above embodiment, for example... Figure 1 The steps S100 to S300 shown, as well as other extensions and related steps of the method, are examples. Alternatively, when the processor executes computer-readable instructions, it implements the functions of each module / unit of the calculation device for the rainstorm warning threshold in the above embodiments, for example... Figure 2 The functions of modules 100 to 300 are shown. To avoid repetition, they will not be described again here.

[0147] A processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of a computer device, connecting all parts of the computer device through various interfaces and lines.

[0148] Memory can be used to store computer-readable instructions and / or modules. The processor implements various functions of the computer device by running or executing the computer-readable instructions and / or modules stored in memory, and by accessing data stored in memory. Memory can mainly include a program storage area and a data storage area. The program storage area can store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area can store data created based on the use of the mobile phone (such as audio data, video data, etc.).

[0149] The memory can be integrated into the processor or set up separately from the processor.

[0150] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0151] In one embodiment, a computer readable storage medium is provided, having computer readable instructions stored thereon, which when executed by a processor implement the steps of the method for calculating the heavy rain warning threshold in the above embodiments, for example Figure 1 implement the functions of the modules / units of the device for calculating the heavy rain warning threshold in the above embodiments, for example Figure 2 the functions of the modules 100 to 300. To avoid repetition, no further elaboration is given here.

[0152] Those skilled in the art can understand that all or part of the processes in the above embodiments can be completed by computer readable instructions instructing the related hardware, and the computer readable instructions can be stored in a computer readable storage medium, and when executed, can include the processes of the above embodiments. Any reference to memory, storage, database or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0153] It should be noted that in this document, the terms "comprising", "including", or any other variant thereof are intended to cover non-exclusive inclusions, so that processes, devices, articles or methods including a series of elements not only include those elements, but also include other elements not explicitly listed, or include elements inherent to such processes, devices, articles or methods. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0154] The above application embodiment serial numbers are only for description, and do not represent the advantages and disadvantages of the embodiments. Through the above description of the embodiments, those skilled in the art can clearly understand that the above embodiment methods can be realized by means of software and the necessary general hardware platform, and of course, they can also be realized by hardware, but in many cases, the former is a better embodiment. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disc, optical disc) as described above, and includes a plurality of instructions for making a terminal device (which can be a mobile phone, computer, server, or network device, etc.) execute the methods described in various embodiments of the present application.

[0155] The above is only the preferred embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation using the content of the present application specification and drawings, or direct or indirect application in other related technical fields, is also included in the patent protection scope of the present application.

Claims

1. A method for calculating a storm warning threshold, characterized in that, The method comprises: obtaining rainfall data of a target region in a preset time period and claim data caused by a storm leading to a risk; calculating a storm risk probability of a storm event and an existing risk event when an estimated storm warning threshold is a different value according to the rainfall data and the claim data, wherein the storm event is an event in which actual rainfall exceeds the corresponding estimated storm warning threshold; determining the estimated storm warning threshold corresponding to the maximum storm risk probability as a final storm warning threshold of the target region; the rainfall data comprises sub-rainfall data of each sub-region of the target region in the preset time period, and the claim data comprises sub-claim data of each sub-region of the target region in the preset time period; the calculating of the storm risk probability of the storm event and the existing risk event when the estimated storm warning threshold is the different value according to the rainfall data and the claim data comprises: calculating a sub-storm risk probability of each sub-region of the storm event and the existing risk event when the estimated storm warning threshold is a same value according to the sub-rainfall data and the sub-claim data; averaging the sub-storm risk probabilities of all target sub-regions corresponding to the estimated storm warning threshold of the same value to obtain a storm risk probability corresponding to the estimated storm warning threshold.

2. The method of claim 1, wherein, the rainfall data comprises sub-rainfall data of each sub-region of the target region in the preset time period, and the claim data comprises sub-claim data of each sub-region of the target region in the preset time period; the calculating of the storm risk probability of the storm event and the existing risk event when the estimated storm warning threshold is the different value according to the rainfall data and the claim data comprises: calculating a risk storm number of each sub-region of the risk event and the existing storm event when the estimated storm warning threshold is the same value according to the sub-rainfall data and the sub-claim data; calculating a storm event number of each sub-region of the storm event when the estimated storm warning threshold is the same value according to the sub-rainfall data; summing the risk storm number of all target sub-regions corresponding to the estimated storm warning threshold of the same value and summing the storm event number of all target sub-regions to calculate a ratio of the sum of the risk storm number to the sum of the storm event number as the storm risk probability corresponding to the estimated storm warning threshold.

3. The method of claim 1, wherein, the calculating of the sub-storm risk probability of each sub-region of the storm event and the existing risk event when the estimated storm warning threshold is the same value according to the sub-rainfall data and the sub-claim data comprises: calculating a risk storm probability of each sub-region of the risk event and the existing storm event when the estimated storm warning threshold is the same value according to the sub-rainfall data and the sub-claim data, and calculating an actual risk probability of the risk event and a storm probability of the storm event; calculating the sub-storm risk probability of each sub-region of the storm event and the existing risk event when the estimated storm warning threshold is the same value according to the risk storm probability, the actual risk probability and the storm probability.

4. The method of claim 3, wherein, The method comprises the following steps of: calculating, according to the sub-rainfall data, the number of rainfall events and the number of heavy rainfall events in each sub-region within the preset time period when the estimated heavy rainfall warning threshold is the same value; calculating, according to the sub-claim data, the number of risk events in each sub-region; calculating, according to the sub-rainfall data and the sub-claim data, the risk and heavy rainfall number of each sub-region when the estimated heavy rainfall warning threshold is the same value; calculating the risk and heavy rainfall probability of the sub-region under the estimated heavy rainfall warning threshold according to the risk and heavy rainfall number and the risk number of the same sub-region when the estimated heavy rainfall warning threshold is the same value; calculating the actual risk probability of the risk event and the heavy rainfall probability of the heavy rainfall event according to the sub-rainfall data and the sub-claim data, comprising: calculating, according to the sub-rainfall data, the number of rainfall events and the number of heavy rainfall events in each sub-region within the preset time period when the estimated heavy rainfall warning threshold is the same value; 5. The method of claim 1, wherein, calculating, according to the sub-claim data, the number of risk events in each sub-region; calculating, according to the sub-rainfall data and the sub-claim data, the risk and heavy rainfall number of each sub-region when the estimated heavy rainfall warning threshold is the same value; calculating the risk and heavy rainfall probability of the sub-region under the estimated heavy rainfall warning threshold according to the risk and heavy rainfall number and the risk number of the same sub-region when the estimated heavy rainfall warning threshold is the same value; calculating the actual risk probability of the risk event and the heavy rainfall probability of the heavy rainfall event according to the sub-rainfall data and the sub-claim data, comprising: calculating, according to the sub-rainfall data, the number of rainfall events and the number of heavy rainfall events in each sub-region within the preset time period when the estimated heavy rainfall warning threshold is the same value; 6. The method of claim 1 or 2, wherein, calculating, according to the sub-claim data, the number of risk events in each sub-region; calculating, according to the sub-rainfall data and the sub-claim data, the risk and heavy rainfall number of each sub-region when the estimated heavy rainfall warning threshold is the same value; calculating the risk and heavy rainfall probability of the sub-region under the estimated heavy rainfall warning threshold according to the risk and heavy rainfall number and the risk number of the same sub-region when the estimated heavy rainfall warning threshold is the same value.

7. A device for computing a thunderstorm warning threshold, the device being configured to implement the method for computing a thunderstorm warning threshold according to any one of claims 1 to 6, characterized in that, The rainfall data comprises the actual daily rainfall in each day within the preset time period; The heavy rainfall event is specifically recorded as one heavy rainfall event when the actual daily rainfall exceeds the corresponding estimated heavy rainfall warning threshold, Or, if there is no continuous multi-day actual daily rainfall exceeding the corresponding estimated heavy rainfall warning threshold, then the event of single-day actual daily rainfall exceeding the corresponding estimated heavy rainfall warning threshold is recorded as one heavy rainfall event, 8. A computer device comprising a memory, a processor, and computer readable instructions stored on the memory and executable on the processor, the computer readable instructions comprising: if there is continuous multi-day actual daily rainfall exceeding the corresponding estimated heavy rainfall warning threshold, then the event of continuous multi-day actual daily rainfall exceeding the corresponding estimated heavy rainfall warning threshold is recorded as one heavy rainfall event. The target sub-region is all sub-regions contained in the target region, Or, the target sub-region is the sub-region remaining after removing the interference region, wherein the interference region is a sub-region with a number of claims less than a minimum claim threshold within the preset time period, or the interference region is a sub-region with a number of insurance less than a minimum insurance threshold within the preset time period. The device comprises: a data acquisition module for acquiring rainfall data and claim data of a target region within a preset time period; a calculation module for calculating the risk and heavy rainfall probability of the risk event and the heavy rainfall event according to the rainfall data and the claim data, wherein the heavy rainfall event is an event in which the actual rainfall exceeds the corresponding estimated heavy rainfall warning threshold; a threshold determination module for determining the estimated heavy rainfall warning threshold corresponding to the maximum risk and heavy rainfall probability as the final heavy rainfall warning threshold of the target region. The processor executes the computer readable instructions to perform the steps of the heavy rainfall warning threshold calculation method of any one of claims 1-6.

9. A computer-readable storage medium having stored thereon computer-readable instructions, wherein, The computer readable instructions, when executed by the processor, cause the processor to perform the steps of the method of calculating a storm warning threshold as claimed in any one of claims 1 to 6.

Images