Search

CN-122022457-A - Method, device, equipment and medium for pushing risk trend data

CN122022457ACN 122022457 ACN122022457 ACN 122022457ACN-122022457-A

Abstract

The application belongs to the field of artificial intelligence, and relates to a method, a device, equipment and a medium for pushing risk trend data. And acquiring monitoring images and meteorological data of the target land crops, and determining the current growing period of the target land crops. And (3) using the growth period data and the meteorological data, evaluating by a meteorological risk evaluation model, and outputting risk degree indexes and risk driving factor analysis results. And correcting the full-period risk threshold according to the growing period to obtain a target risk threshold. And acquiring weather forecast data of a target time period, and forecasting by adopting a time sequence forecast model in combination with the result to obtain risk trend data. And extracting user information of the target land block from the crop multisource data, and pushing the risk trend data to the corresponding user terminal. The application can be applied to the business fields of financial insurance, agricultural management and the like, and can improve the accuracy of pushing the crop risk trend data.

Inventors

  • HE JI

Assignees

  • 中国平安财产保险股份有限公司

Dates

Publication Date
20260512
Application Date
20260114

Claims (10)

  1. 1. The pushing method of the risk trend data is characterized by comprising the following steps: Acquiring crop multisource data from a preset industry knowledge multisource database; Coupling the crop multisource data with an initial weather prediction model to obtain a weather risk assessment model; acquiring a current crop monitoring image and meteorological data of a target land block, and determining a target growth period of the current target crop of the target land block based on the monitoring image; based on the growth period data of the target growth period and the meteorological data, adopting the meteorological risk assessment model to carry out dynamic risk assessment, and outputting risk degree indexes and risk driving factor analysis results of the target crops; Correcting a preset full-period risk threshold based on the target growing period to obtain a target risk threshold; acquiring weather forecast data of a target time period, and carrying out risk trend forecast by adopting a time sequence forecast model based on the risk degree index, the risk driving factor analysis result, the target risk threshold value and the weather forecast data to obtain risk trend data of the target crop in the target time period; And acquiring user information of the target land block from the crop multisource data, and pushing the risk trend data to a user terminal corresponding to the user information.
  2. 2. The method of claim 1, wherein the crop multisource data comprises industry knowledge graph, crop space-time correlation data, insurance business data, user feedback data, historical meteorological data, and wherein the step of coupling the crop multisource data with an initial meteorological prediction model to obtain a meteorological risk assessment model comprises the following steps: carrying out multi-source data fusion on the historical meteorological data, the industry knowledge graph, crop space-time associated data, insurance business data, user feedback data and the historical meteorological data to obtain a multi-dimensional fusion data set; Extracting characteristic information related to weather risks from the multi-dimensional fusion data set; And carrying out parameter adjustment on the initial weather forecast model by adopting the characteristic information, and updating the characteristic weight and the risk level judgment boundary to obtain the weather risk assessment model.
  3. 3. The method according to any one of claims 1 or 2, wherein the step of performing dynamic risk assessment using the weather risk assessment model based on the growth period data of the target growth period and the weather data, and outputting risk level indicators and risk driving factor analysis results of the target crop specifically includes: according to the growth period data and the meteorological data, adopting the meteorological risk assessment model to carry out dynamic risk assessment to obtain a risk degree index of the target crop; extracting a plurality of key meteorological parameters affecting the risk degree from the meteorological data based on the characteristic weight of the meteorological risk assessment model; based on the multiple key meteorological parameters and the risk sensitive characteristics of the target growing period, combining a risk grade judgment boundary of the meteorological risk assessment model, screening out parameters exceeding the corresponding grade boundary, and determining risk driving factors; and if the risk degree index exceeds a preset threshold value, generating a corresponding risk driving factor analysis result based on the association rule of the risk driving factor and the target growing period.
  4. 4. The method according to claim 1, wherein the step of determining the target growth period in which the target crop of the target plot is currently located based on the monitoring image, specifically comprises: extracting image features of the target crop according to the monitoring image, wherein the image features comprise morphological features and color features; Analyzing the morphological characteristics and the color characteristics by adopting a pre-established crop growth model to obtain growth parameters of target crops; comparing the growth parameters with preset growth period judging conditions to obtain a comparison result; And determining the current target growing period of the target crop based on the comparison result.
  5. 5. The method according to claim 1, wherein the full-cycle risk threshold comprises a preset temperature threshold and a preset humidity threshold, and wherein the step of correcting the preset full-cycle risk threshold based on the target growth period to obtain the target risk threshold comprises the following steps: Acquiring a temperature sensitivity coefficient and a humidity sensitivity coefficient of the target crop to a risk factor in the target growth period from the industry knowledge multisource database; calculating the ratio between the preset temperature threshold and the temperature sensitivity coefficient, and determining the ratio as a target temperature threshold; calculating the product between the preset humidity threshold and the humidity sensitivity coefficient, and determining the product as a target humidity threshold; Determining quantization thresholds corresponding to different risk levels based on the target temperature threshold and the target humidity threshold; and determining the target temperature threshold, the target humidity threshold and the quantification threshold as target risk thresholds.
  6. 6. The method according to claim 5, wherein the step of predicting a risk trend using a time-series prediction model based on the risk level indicator, the risk driving factor analysis result, the target risk threshold, and the weather prediction data to obtain risk trend data of the target crop in the target time period specifically includes: aligning and associating the risk degree index, the risk driving factor analysis result, the target risk threshold value and the weather forecast data according to time sequence to obtain a time sequence data set; Extracting risk trend associated features from the time series data set; Inputting the risk trend associated features into a time sequence prediction model to obtain a daily risk degree prediction value in a target time period; comparing the daily risk degree predicted value with the quantization threshold value, and judging a daily risk level; Generating a risk trend table and a risk trend graph based on the daily risk level and key driving factors of the weather forecast data; And determining the risk trend table and the risk trend graph as risk trend data of the target crop in the target time period.
  7. 7. The method of claim 6, wherein the step of pushing the risk trend data to the user terminal corresponding to the user information specifically includes: Converting the risk trend table into a universal format file; The risk trend graph is exported to be a risk trend image, and a risk abstract text is generated based on the risk trend image; checking the terminal identification format and the registration state in the user information, and constructing a corresponding relation table of the user identification and the terminal identification; determining an adaptive pushing channel based on the terminal type in the corresponding relation table; And pushing the risk abstract text and the universal format file to the user terminal corresponding to the user identifier through the pushing channel.
  8. 8. A pushing device for risk trend data, comprising: the acquisition module is used for acquiring crop multisource data from a preset industry knowledge multisource database; the coupling module is used for coupling the crop multisource data with the initial weather prediction model to obtain a weather risk assessment model; The determining module is used for acquiring a current crop monitoring image and meteorological data of a target land block, and determining a target growing period of the target crop of the target land block on the basis of the monitoring image; the assessment module is used for carrying out dynamic risk assessment by adopting the weather risk assessment model based on the growth period data of the target growth period and the weather data, and outputting risk degree indexes and risk driving factor analysis results of the target crops; The correction module is used for correcting a preset full-period risk threshold value based on the target growing period to obtain a target risk threshold value; the prediction module is used for acquiring weather prediction data of a target time period, and performing risk trend prediction by adopting a time sequence prediction model based on the risk degree index, the risk driving factor analysis result, the target risk threshold value and the weather prediction data to obtain risk trend data of the target crop in the target time period; And the pushing module is used for acquiring the user information of the target land block from the crop multisource data and pushing the risk trend data to a user terminal corresponding to the user information.
  9. 9. A computer device comprising a memory and a processor, the memory having stored therein computer readable instructions which when executed by the processor implement the steps of the method of pushing risk trend data according to any one of claims 1 to 7.
  10. 10. A computer readable storage medium, wherein computer readable instructions are stored on the computer readable storage medium, which when executed by a processor, implement the steps of the method for pushing risk trend data according to any one of claims 1 to 7.

Description

Method, device, equipment and medium for pushing risk trend data Technical Field The application relates to the technical field of artificial intelligence, is applied to online processing of business scenes of financial insurance, agricultural management and the like, and particularly relates to a pushing method, device, equipment and medium of risk trend data. Background Under the background of frequent global extreme climate and aggravated natural disaster risk, the intelligent upgrading of the disaster prevention and reduction system is not delayed, and the evolution of artificial intelligence from 'passive response' to 'active defense' is a key trend for pushing the disaster prevention and reduction work to a new height. However, in the prior art in the field of disaster reduction in the agricultural insurance industry, a plurality of short plates exist, so that the improvement of the actual disaster prevention and reduction effect is restricted. The conventional AI disaster reduction system has obvious problems in weather prediction and risk assessment. Most systems adopt general weather model prediction, and are not fully integrated with business logic and space-time characteristics of specific industries such as agriculture and the like. For example, the difference of the sensitivity of different crops to disasters is large in each growth period, frost is afraid in winter wheat jointing period, and the grouting period is easily influenced by dry hot air, but the existing system only judges risks based on static threshold values, ignores the key variable of the crop growth period, and causes early warning and actual damage disconnection. In addition, the system interaction mode is backward, mainly comprises 'passive question and answer', lacks the capability of active guidance and intelligent recommendation, excessively depends on active inquiry of a user, cannot deduce active prompt risk based on trend, has lagged perception of the risk of the user, and is difficult to prevent in advance. Even if part of the system is accessed into a language large model to support multiple rounds of conversations, the nature is still responsive interaction, and a real active communication mechanism is not constructed. Disclosure of Invention The embodiment of the application aims to provide a method, a device, computer equipment and a storage medium for pushing risk trend data, which are used for solving the problems that the existing disaster reduction system does not combine agriculture business logic and space-time characteristics, and a manual interaction mode is behind, active communication is lacking, so that risk early warning and prevention effects are poor. In a first aspect, a method for pushing risk trend data is provided, and the following technical scheme is adopted: The method comprises the steps of obtaining crop multisource data from a preset industry knowledge multisource database, coupling the crop multisource data with an initial weather prediction model to obtain a weather risk assessment model, obtaining a current crop monitoring image and weather data of a target land block, determining a target growing period of the target crop of the target land block on the basis of the monitoring image, carrying out dynamic risk assessment by adopting a weather risk assessment model on the basis of growth period data and weather data of the target growing period, outputting risk degree indexes and risk driving factor analysis results of the target crop, correcting a preset full-period risk threshold on the basis of the target growing period to obtain a target risk threshold, obtaining weather prediction data of a target time period, carrying out risk trend prediction by adopting a time sequence prediction model on the basis of the risk degree indexes, the risk driving factor analysis results, the target risk threshold and the weather prediction data, obtaining risk trend data of the target land block in the target time period, obtaining user information of the target land block from the crop multisource data, and pushing the risk trend data to a user terminal corresponding to the user information. In a second aspect, a device for pushing risk trend data is provided, which adopts the following technical scheme: the acquisition module is used for acquiring crop multisource data from a preset industry knowledge multisource database; The coupling module is used for coupling the crop multisource data with the initial weather prediction model to obtain a weather risk assessment model; The determining module is used for acquiring a current crop monitoring image and meteorological data of the target land block and determining a target growing period of the target crop of the target land block based on the monitoring image; The assessment module is used for carrying out dynamic risk assessment by adopting a weather risk assessment model based on the growth period data and the weather data of the target growth period