A multi-source meteorological data integration and intelligent analysis system

Abstract

The application discloses a kind of multi-source meteorological data integration and intelligent analysis system, comprising: establishing module, for establishing the several meteorological indexes of multi-source meteorological data, obtain the historical information of meteorological index, according to historical information set the priority coefficient of corresponding meteorological index;Generation module, for generating meteorological index evaluation model according to meteorological index, and establish the corresponding relationship between meteorological index and meteorological index evaluation model;Evaluation module, for calculating the meteorological index of meteorological data in the preset period, based on meteorological index evaluation model and priority coefficient, determine the index evaluation value of each meteorological index of current meteorological data;Visual module, for generating the forecast result of corresponding meteorological data and visual display according to the index evaluation value of each meteorological index of the same meteorological data. By improving the evaluation accuracy of meteorological data forecast, meteorological data meeting user demand can be quickly provided, and the accuracy and timeliness of new energy market transaction decision-making are improved.

Description

Technical Field

[0001] This invention relates to the field of meteorological data analysis technology, and in particular to a multi-source meteorological data integration and intelligent analysis system. Background Technology

[0002] New energy power forecasting involves inputting meteorological forecasts into a power generation model to calculate how much electricity wind and solar power can generate. Based on this, a production scheduling plan is formed to ensure proper operation and power transmission. Therefore, meteorological data forecasts are one of the most important factors affecting the accuracy of new energy power forecasting.

[0003] In existing technologies, most methods use a single meteorological calculation indicator to evaluate the accuracy of meteorological data. These methods are simple but lack comprehensiveness, which reduces the accuracy of meteorological data forecasts. Furthermore, the large volume of meteorological data processing makes it impossible to quickly provide meteorological data that meets user needs, thereby reducing the accuracy and timeliness of new energy market transaction decisions. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a multi-source meteorological data integration and intelligent analysis system. This system integrates and processes multi-source meteorological data, establishes multiple meteorological indicators, constructs multiple meteorological indicator evaluation models, and obtains the evaluation value of each meteorological indicator based on the evaluation models and priority coefficients. Based on the evaluation values ​​of multiple meteorological indicators for the same meteorological data, the system obtains the current meteorological data forecast result and displays it visually, thereby improving the accuracy of meteorological data forecast evaluation. It can quickly provide meteorological data that meets user needs, improving the accuracy and timeliness of new energy market transaction decisions.

[0005] This invention provides a multi-source meteorological data integration and intelligent analysis system, comprising:

[0006] The module is used to establish several meteorological indicators from multi-source meteorological data, obtain historical information of meteorological indicators, and set priority coefficients for corresponding meteorological indicators based on historical information.

[0007] The generation module is used to generate meteorological indicator evaluation models based on meteorological indicators and to establish the correspondence between meteorological indicators and meteorological indicator evaluation models.

[0008] The evaluation module is used to calculate meteorological indicators of meteorological data for a preset time period. Based on the meteorological indicator evaluation model and priority coefficient, it determines the indicator evaluation value of each meteorological indicator of the current meteorological data.

[0009] The visualization module is used to generate forecast results for the corresponding meteorological data based on the indicator evaluation values ​​of each meteorological indicator of the same meteorological data, and to visualize the forecast results.

[0010] In some embodiments of the present invention, the meteorological indicators include the root mean square error, completeness, effectiveness, correlation, and MAE of multi-source meteorological data forecasts.

[0011] In some embodiments of the present invention, priority coefficients for corresponding meteorological indicators are set based on historical information, including:

[0012] Obtain historical information on meteorological indicators, including generation identifiers, related identifiers, and application identifiers;

[0013] The initial acquisition form and final generation form of the current meteorological indicator are determined based on the generation identifier of the meteorological indicator. According to the initial acquisition form and final generation form, the relevant formulas are retrieved from the identifier-formula database, and the relevant formulas are simulated to obtain the calculation time of each relevant formula. The generation framework of the corresponding meteorological indicator is obtained according to the calculation order and calculation time of the relevant formulas. The generation time of the corresponding meteorological indicator is set according to the route length of the generation framework.

[0014] Based on multiple completion markers of meteorological indicators, the historical meteorological indicator change curves corresponding to the multiple completion markers are obtained. Based on the multiple historical meteorological indicator change curves and the corresponding historical forecast results, the average forecast rate of the historical forecast results of the historical meteorological indicator change curves is generated. Based on the forecast success rate threshold, the number of abnormal historical meteorological indicator change curves is screened out, and the forecast stability of the current meteorological indicator is generated.

[0015] The first and second application frequencies are determined based on the application identifiers of meteorological indicators. The first influence weight of the number of times the corresponding meteorological indicator is used alone is obtained based on the first application frequency. The second influence weight of the number of times the corresponding meteorological indicator is used in combination is obtained based on the second application frequency. The comprehensive influence weight of the corresponding meteorological indicator is generated based on the first and second influence weights.

[0016] Priority coefficients for corresponding meteorological indicators are generated based on their generation time, forecast stability, and overall impact weight.

[0017] In some embodiments of the present invention, priority coefficients for corresponding meteorological indicators are generated based on the generation time, forecast stability, and comprehensive impact weight of the meteorological indicators, including:

[0018] The formula for calculating the priority coefficient is as follows:

[0019] K=a1*e1*t+a2*e2*W+a3*e3*(u1*r1+u2*r2) / n;

[0020] Where K is the priority coefficient, a1 is the generation time conversion coefficient, e1 is the weight coefficient corresponding to the generation time, t is the generation time, a2 is the forecast stability conversion coefficient, e2 is the weight coefficient corresponding to the forecast stability, W is the forecast stability, a3 is the comprehensive influence weight conversion coefficient, e3 is the weight coefficient corresponding to the comprehensive influence weight, (u1*r1+u2*r2) / n is the comprehensive influence weight, u1 is the first influence weight, r1 is the weight coefficient of the first influence weight, u2 is the second influence weight, r2 is the weight coefficient of the second influence weight, and n is the combined number of times used alone and used in combination.

[0021] Based on the relationship between the priority coefficients of multiple meteorological indicators of the same meteorological data and the preset priority coefficient threshold, set the label information of the corresponding meteorological indicators;

[0022] Pre-set a first preset preference coefficient threshold and a second preset priority coefficient threshold;

[0023] When the priority coefficient is at the first preset priority coefficient threshold, an ignore label is set for the corresponding meteorological indicator;

[0024] When the priority coefficient is at the first preset priority coefficient threshold and the second preset priority coefficient threshold, a secondary label is set for the corresponding weather indicator;

[0025] When the priority coefficient is greater than the second preset priority coefficient threshold, an important label is set for the corresponding meteorological indicator and cached.

[0026] In some embodiments of the present invention, generating a meteorological indicator evaluation model based on meteorological indicators includes:

[0027] Obtain the relevant data and historical evaluation values ​​of each meteorological indicator for historical periods;

[0028] Training and test set data are generated based on relevant indicator data and historical indicator evaluation values.

[0029] A meteorological index evaluation model is generated based on the training set data, and the credibility of the meteorological index evaluation model is generated based on the test set data.

[0030] Preset credibility threshold;

[0031] If the credibility of the meteorological index evaluation model is less than the credibility threshold, the training set data is used for iterative training, and the meteorological index evaluation model is regenerated until the credibility is greater than the credibility threshold, at which point the iterative training stops.

[0032] If the credibility of the meteorological indicator evaluation model is greater than the credibility threshold, an index label for the meteorological indicator evaluation model is generated.

[0033] Establish a meteorological indicator-index label mapping table, and generate the correspondence between the meteorological indicator evaluation model and the meteorological indicators based on the meteorological indicator-index label mapping table.

[0034] In some embodiments of the present invention, the evaluation value of each meteorological indicator in the current meteorological data is determined based on a meteorological indicator evaluation model and a priority coefficient, including:

[0035] Obtain meteorological indicators from the current meteorological data, and filter out the meteorological indicator evaluation model corresponding to the current meteorological indicator based on the meteorological indicator-index label mapping table;

[0036] Input the relevant data of the meteorological indicators for the current preset period into the corresponding meteorological indicator evaluation model to obtain the first indicator evaluation value of the current meteorological indicator;

[0037] Based on the evaluation value of the first indicator and the priority coefficient of the corresponding meteorological indicator, the evaluation value G of the current meteorological indicator is generated.

[0038] G = g1 * K;

[0039] Among them, g1 is the first indicator evaluation value of the current meteorological indicators.

[0040] In some embodiments of the present invention, a forecast result for the corresponding meteorological data is generated based on the index evaluation value of each meteorological index of the same meteorological data, including:

[0041] The evaluation values ​​of each meteorological indicator for the same meteorological data are averaged to obtain the evaluation mean of the corresponding meteorological data.

[0042] A first preset evaluation mean interval, a second preset evaluation mean interval, and a third preset evaluation mean interval are pre-defined;

[0043] When the mean value of the meteorological data is within the first preset mean value range, the forecast result of the current meteorological data is an incorrect prediction.

[0044] When the average value of the meteorological data is within the second preset average value range, the forecast result of the current meteorological data is a deviation prediction.

[0045] When the average value of the meteorological data is within the third preset average value range, the forecast result of the current meteorological data is an accurate prediction.

[0046] In some embodiments of the present invention, before generating the forecast result corresponding to the meteorological data, the method further includes:

[0047] Obtain the configuration requirements from the user's end, and based on the configuration requirements, filter out specific meteorological data and set the configuration parameters of the indicator configuration components for specific meteorological data;

[0048] The indicator evaluation value is generated based on the indicator configuration component under the corresponding configuration parameters of specific meteorological data, and then visualized according to the preset display format.

[0049] The multi-source meteorological data integration and intelligent analysis system provided by this invention has the following advantages compared with the prior art:

[0050] By integrating and processing multi-source meteorological data and establishing multiple meteorological indicators, multiple meteorological indicator evaluation models are constructed. Based on the meteorological indicator evaluation models and priority coefficients, the indicator evaluation value of each meteorological indicator is obtained. Based on the indicator evaluation values ​​of multiple meteorological indicators of the same meteorological data, the current meteorological data forecast result is obtained and visualized, which improves the evaluation accuracy of meteorological data forecasts. It can quickly provide meteorological data that meets user needs and improve the accuracy and timeliness of new energy market transaction decisions. Attached Figure Description

[0051] Figure 1 This is a schematic diagram of a multi-source meteorological data integration and intelligent analysis system in a preferred embodiment of the present invention. Detailed Implementation

[0052] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0053] In the description of this invention, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0054] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0055] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0056] like Figure 1 As shown, a preferred embodiment of the multi-source meteorological data integration and intelligent analysis system of the present invention includes:

[0057] The module is used to establish several meteorological indicators from multi-source meteorological data, obtain historical information of meteorological indicators, and set priority coefficients for corresponding meteorological indicators based on historical information.

[0058] The generation module is used to generate meteorological indicator evaluation models based on meteorological indicators and to establish the correspondence between meteorological indicators and meteorological indicator evaluation models.

[0059] The evaluation module is used to calculate meteorological indicators of meteorological data for a preset time period. Based on the meteorological indicator evaluation model and priority coefficient, it determines the indicator evaluation value of each meteorological indicator of the current meteorological data.

[0060] The visualization module is used to generate forecast results for the corresponding meteorological data based on the indicator evaluation values ​​of each meteorological indicator of the same meteorological data, and to visualize the forecast results.

[0061] In this embodiment, the system accesses multi-source meteorological data with different temporal and spatial resolutions, and supports various formats of meteorological data and various access methods, including temporal resolution, spatial resolution, meteorological file format, access method, etc., to achieve integrated processing of multi-source meteorological data.

[0062] In this embodiment, several meteorological indicators from multiple sources are established based on historical meteorological data from the past two years. Priority coefficients for the corresponding meteorological indicators are set based on the historical information of the meteorological indicators from the past two years. The meteorological data for the preset time period refers to the meteorological data within one week. The real-time meteorological data for each day is forecasted based on the meteorological data within one week. The forecast results are obtained based on the indicator evaluation values ​​of multiple meteorological indicators from the meteorological data. The multi-source meteorological data is comprehensively analyzed, and meteorological data with high prediction accuracy is selected to improve the accuracy of new energy power prediction.

[0063] In this embodiment, the calculation of the evaluation value of each meteorological indicator is based on the priority coefficient of each meteorological indicator. The important indicators of multiple meteorological indicators are selected according to the priority coefficient. The important indicators are calculated and cached in advance. When the user retrieves the corresponding meteorological indicator, the cached content can be displayed directly, thereby improving the accuracy and timeliness of new energy market transaction decisions.

[0064] In some embodiments of the present invention, the meteorological indicators include indicators such as root mean square error, completeness, effectiveness, correlation, and MAE (average absolute error) of multi-source meteorological data forecasts.

[0065] In some embodiments of the present invention, priority coefficients for corresponding meteorological indicators are set based on historical information, including:

[0066] Obtain historical information on meteorological indicators, including generation identifiers, related identifiers, and application identifiers;

[0067] The initial acquisition form and final generation form of the current meteorological indicator are determined based on the generation identifier of the meteorological indicator. According to the initial acquisition form and final generation form, the relevant formulas are retrieved from the identifier-formula database, and the relevant formulas are simulated to obtain the calculation time of each relevant formula. The generation framework of the corresponding meteorological indicator is obtained according to the calculation order and calculation time of the relevant formulas. The generation time of the corresponding meteorological indicator is set according to the route length of the generation framework.

[0068] Based on multiple completion markers of meteorological indicators, the historical meteorological indicator change curves corresponding to the multiple completion markers are obtained. Based on the multiple historical meteorological indicator change curves and the corresponding historical forecast results, the average forecast rate of the historical forecast results of the historical meteorological indicator change curves is generated. Based on the forecast success rate threshold, the number of abnormal historical meteorological indicator change curves is screened out, and the forecast stability of the current meteorological indicator is generated.

[0069] The first and second application frequencies are determined based on the application identifiers of meteorological indicators. The first influence weight of the number of times the corresponding meteorological indicator is used alone is obtained based on the first application frequency. The second influence weight of the number of times the corresponding meteorological indicator is used in combination is obtained based on the second application frequency. The comprehensive influence weight of the corresponding meteorological indicator is generated based on the first and second influence weights.

[0070] Priority coefficients for corresponding meteorological indicators are generated based on their generation time, forecast stability, and overall impact weight.

[0071] In this embodiment, the generation identifier of each meteorological indicator is different. The generation identifier refers to the one calculated by each meteorological indicator using at least one generation formula. The completion identifier refers to the identifier when the meteorological indicator calculation is completed. Based on the meteorological indicator data corresponding to the completion of the calculation, the corresponding historical meteorological indicator change curve is generated. Based on the relationship between the historical meteorological indicator change curve and the historical forecast results, a meteorological indicator change curve-forecast rate mapping table is constructed to filter out the average forecast rate corresponding to the current meteorological indicator, thereby filtering out the number of abnormal historical meteorological indicator change curves. The normal historical meteorological indicator change curve is the curve whose historical forecast result is greater than the average forecast rate. 1 - number of abnormal historical meteorological indicator change curves / number of normal historical meteorological indicator change curves = forecast stability.

[0072] In this embodiment, the first application frequency refers to the degree of influence of the corresponding meteorological indicator on the forecast results of the corresponding meteorological data when the corresponding meteorological indicator is analyzed alone (i.e., the first influence weight), and the second application frequency refers to the degree of influence of the corresponding meteorological indicator on the forecast results of the corresponding meteorological data when the corresponding meteorological indicator is analyzed together with other meteorological indicators (i.e., the second influence weight).

[0073] In this embodiment, priority coefficients for meteorological indicators are generated based on their generation time, forecast stability, and comprehensive impact weight. This lays the data foundation for subsequent calculation of indicator evaluation values. To facilitate timely scheduling of corresponding meteorological indicators by users, meteorological indicators with high priority coefficients are cached first, thereby ensuring user experience and improving the efficiency of meteorological data utilization.

[0074] In some embodiments of the present invention, priority coefficients for corresponding meteorological indicators are generated based on the generation time, forecast stability, and comprehensive impact weight of the meteorological indicators, including:

[0075] The formula for calculating the priority coefficient is as follows:

[0076] K=a1*e1*t+a2*e2*W+a3*e3*(u1*r1+u2*r2) / n;

[0077] Where K is the priority coefficient, a1 is the generation time conversion coefficient, e1 is the weight coefficient corresponding to the generation time, t is the generation time, a2 is the forecast stability conversion coefficient, e2 is the weight coefficient corresponding to the forecast stability, W is the forecast stability, a3 is the comprehensive influence weight conversion coefficient, e3 is the weight coefficient corresponding to the comprehensive influence weight, (u1*r1+u2*r2) / n is the comprehensive influence weight, u1 is the first influence weight, r1 is the weight coefficient of the first influence weight, u2 is the second influence weight, r2 is the weight coefficient of the second influence weight, and n is the combined number of times used alone and used in combination.

[0078] Based on the relationship between the priority coefficients of multiple meteorological indicators of the same meteorological data and the preset priority coefficient threshold, set the label information of the corresponding meteorological indicators;

[0079] Pre-set a first preset preference coefficient threshold and a second preset priority coefficient threshold;

[0080] When the priority coefficient is at the first preset priority coefficient threshold, an ignore label is set for the corresponding meteorological indicator;

[0081] When the priority coefficient is at the first preset priority coefficient threshold and the second preset priority coefficient threshold, a secondary label is set for the corresponding weather indicator;

[0082] When the priority coefficient is greater than the second preset priority coefficient threshold, an important label is set for the corresponding meteorological indicator and cached.

[0083] In this embodiment, corresponding tag information is set according to the priority coefficient of meteorological indicators, including important, minor and ignore tags. After the user selects meteorological data, the scheduling order of meteorological indicators can be set according to the tag information, and meteorological indicators with important tags can be directly viewed in the cache library, which greatly improves the user's viewing convenience and the efficiency of meteorological data use.

[0084] In some embodiments of the present invention, generating a meteorological indicator evaluation model based on meteorological indicators includes:

[0085] Obtain the relevant data and historical evaluation values ​​of each meteorological indicator for historical periods;

[0086] Training and test set data are generated based on relevant indicator data and historical indicator evaluation values.

[0087] A meteorological index evaluation model is generated based on the training set data, and the credibility of the meteorological index evaluation model is generated based on the test set data.

[0088] Preset credibility threshold;

[0089] If the credibility of the meteorological index evaluation model is less than the credibility threshold, the training set data is used for iterative training, and the meteorological index evaluation model is regenerated until the credibility is greater than the credibility threshold, at which point the iterative training stops.

[0090] If the credibility of the meteorological indicator evaluation model is greater than the credibility threshold, an index label for the meteorological indicator evaluation model is generated.

[0091] Establish a meteorological indicator-index label mapping table, and generate the correspondence between the meteorological indicator evaluation model and the meteorological indicators based on the meteorological indicator-index label mapping table.

[0092] In this embodiment, the indicator-related data of meteorological indicators refers to the measurement data associated with the indicators. For example, when the meteorological indicator is the root mean square error, the indicator-related data includes the deviation between the measured value and the true value in the historical period. The historical indicator evaluation value is the evaluation of the meteorological forecast performance based on the indicator-related data. When the historical indicator evaluation value is larger, it indicates that the meteorological forecast of the current indicator-related data is more accurate.

[0093] In some embodiments of the present invention, the evaluation value of each meteorological indicator in the current meteorological data is determined based on a meteorological indicator evaluation model and a priority coefficient, including:

[0094] Obtain meteorological indicators from the current meteorological data, and filter out the meteorological indicator evaluation model corresponding to the current meteorological indicator based on the meteorological indicator-index label mapping table;

[0095] Input the relevant data of the meteorological indicators for the current preset period into the corresponding meteorological indicator evaluation model to obtain the first indicator evaluation value of the current meteorological indicator;

[0096] Based on the evaluation value of the first indicator and the priority coefficient of the corresponding meteorological indicator, the evaluation value G of the current meteorological indicator is generated.

[0097] G = g1 * K;

[0098] Among them, g1 is the first indicator evaluation value of the current meteorological indicators.

[0099] In some embodiments of the present invention, a forecast result for the corresponding meteorological data is generated based on the index evaluation value of each meteorological index of the same meteorological data, including:

[0100] The evaluation values ​​of each meteorological indicator for the same meteorological data are averaged to obtain the evaluation mean of the corresponding meteorological data.

[0101] A first preset evaluation mean interval, a second preset evaluation mean interval, and a third preset evaluation mean interval are pre-defined;

[0102] When the mean value of the meteorological data is within the first preset mean value range, the forecast result of the current meteorological data is an incorrect prediction.

[0103] When the average value of the meteorological data is within the second preset average value range, the forecast result of the current meteorological data is a deviation prediction.

[0104] When the average value of the meteorological data is within the third preset average value range, the forecast result of the current meteorological data is an accurate prediction.

[0105] In this embodiment, the first preset evaluation mean interval < the second preset evaluation mean interval < the third preset evaluation mean interval. The preset evaluation mean interval is set in advance based on the historical evaluation mean and historical forecast results. The more accurate the forecast result, the larger the evaluation mean should be. Incorrect prediction means that the forecast result accounts for 30% of the correct result, biased prediction means that the forecast result accounts for 60% of the correct result, and accurate prediction means that the forecast result accounts for 90% or more of the correct result.

[0106] In this embodiment, the average value of the evaluation is obtained based on the evaluation values ​​of multiple meteorological indicators, thereby improving the accuracy of the evaluation of meteorological data forecasts. This allows for the accurate determination of the performance of meteorological data forecasts based on meteorological indicators, the selection of meteorological data with high forecast accuracy, and the improvement of the accuracy of new energy power prediction. Consequently, the accuracy and timeliness of new energy market transaction decisions are enhanced.

[0107] In some embodiments of the present invention, before generating the forecast result corresponding to the meteorological data, the method further includes:

[0108] Obtain the configuration requirements from the user's end, and based on the configuration requirements, filter out specific meteorological data and set the configuration parameters of the indicator configuration components for specific meteorological data;

[0109] The indicator evaluation value is generated based on the indicator configuration component under the corresponding configuration parameters of specific meteorological data, and then visualized according to the preset display format.

[0110] In this embodiment, the configuration requirements of the user end include date ranges and specific meteorological data, as well as specific scenario requirements or custom meteorological indicators for the specific meteorological data. Specific scenario requirements include periods of strong and weak winds, peak and trough periods of wind speed, and wind speed magnitude, etc. The indicator configuration component refers to the support for custom configuration settings of various indicators, such as forecast days settings, time period settings, and threshold settings, etc. The user end can select and configure the required meteorological indicators through a visual interface, providing an indicator configuration interface that allows users to add, delete, sort, and combine indicators from the list of available meteorological indicators. Users can define the display format of meteorological indicators, including line charts, bar charts, pie charts, etc., and configure attributes such as colors, labels, and units.

[0111] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and substitutions can be made without departing from the technical principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present invention.

Claims

1. A multi-source meteorological data integration and intelligent analysis system, characterized in that, include: The module is used to establish several meteorological indicators from multi-source meteorological data, obtain historical information of meteorological indicators, and set priority coefficients for corresponding meteorological indicators based on historical information. The generation module is used to generate meteorological indicator evaluation models based on meteorological indicators and to establish the correspondence between meteorological indicators and meteorological indicator evaluation models. The evaluation module is used to calculate meteorological indicators of meteorological data for a preset time period. Based on the meteorological indicator evaluation model and priority coefficient, it determines the indicator evaluation value of each meteorological indicator of the current meteorological data. The visualization module is used to generate forecast results for the corresponding meteorological data based on the index evaluation value of each meteorological indicator of the same meteorological data, and to visualize the forecast results. Priority coefficients for corresponding meteorological indicators are set based on historical information, including: Obtain historical information on meteorological indicators, including generation identifiers, related identifiers, and application identifiers; The initial acquisition form and final generation form of the current meteorological indicator are determined based on the generation identifier of the meteorological indicator. According to the initial acquisition form and final generation form, the relevant formulas are retrieved from the identifier-formula database, and the relevant formulas are simulated to obtain the calculation time of each relevant formula. The generation framework of the corresponding meteorological indicator is obtained according to the calculation order and calculation time of the relevant formulas. The generation time of the corresponding meteorological indicator is set according to the route length of the generation framework. Based on multiple completion markers of meteorological indicators, the historical meteorological indicator change curves corresponding to the multiple completion markers are obtained. Based on the multiple historical meteorological indicator change curves and the corresponding historical forecast results, the average forecast rate of the historical forecast results of the historical meteorological indicator change curves is generated. Based on the forecast success rate threshold, the number of abnormal historical meteorological indicator change curves is screened out, and the forecast stability of the current meteorological indicator is generated. The first and second application frequencies are determined based on the application identifiers of meteorological indicators. The first influence weight of the number of times the corresponding meteorological indicator is used alone is obtained based on the first application frequency. The second influence weight of the number of times the corresponding meteorological indicator is used in combination is obtained based on the second application frequency. The comprehensive influence weight of the corresponding meteorological indicator is generated based on the first and second influence weights. Priority coefficients for corresponding meteorological indicators are generated based on their generation time, forecast stability, and overall impact weight.

2. The multi-source weather data integration and intelligent analysis system of claim 1, wherein, The meteorological indicators include the root mean square error, completeness, effectiveness, correlation, and MAE of multi-source meteorological data forecasts.

3. The multi-source weather data integration and intelligent analysis system of claim 2, wherein, Priority coefficients for corresponding meteorological indicators are generated based on their generation time, forecast stability, and comprehensive impact weights, including: The formula for calculating the priority coefficient is as follows: K=a1*e1*t+a2*e2*W+a3*e3*(u1*r1+u2*r2) / n; Wherein, K is a priority coefficient, a1 is a generation time conversion coefficient, e1 is a weight coefficient corresponding to the generation time, t is the generation time, a2 is a prediction stability conversion coefficient, e2 is a weight coefficient corresponding to the prediction stability, W is the prediction stability, a3 is a comprehensive influence weight conversion coefficient, e3 is a weight coefficient corresponding to the comprehensive influence weight, (u1*r1+u2*r2) / n is the comprehensive influence weight, u1 is the first influence weight, r1 is the weight coefficient of the first influence weight, u2 is the second influence weight, r2 is the weight coefficient of the second influence weight, and n is the comprehensive number of the single use number and the joint use number; According to the relationship between the priority coefficient of each meteorological index of the same meteorological data and the preset priority coefficient threshold, the label information of the corresponding meteorological index is set; The first preset priority coefficient threshold and the second preset priority coefficient threshold are preset; When the priority coefficient is in the first preset priority coefficient threshold, the corresponding meteorological index is set to ignore label; When the priority coefficient is in the first preset priority coefficient threshold and the second preset priority coefficient threshold, the corresponding weather index is set to secondary label; When the priority coefficient is greater than the second preset priority coefficient threshold, the corresponding meteorological index is set to important label, and the cache processing is performed.

4. The multi-source weather data integration and intelligent analysis system of claim 3, wherein, According to the meteorological index, a meteorological index evaluation model is generated, including: Obtaining the index related data and the historical index evaluation value of each meteorological index in the historical period; According to the index related data and the historical index evaluation value, training set data and test set data are generated; According to the training set data, the meteorological index evaluation model is generated, and according to the test set data, the credibility of the meteorological index evaluation model is generated; A credibility threshold is preset; If the credibility of the meteorological index evaluation model is less than the credibility threshold, the training set data is iteratively trained, and the meteorological index evaluation model is regenerated until the credibility is greater than the credibility threshold, and the iterative training is stopped; If the credibility of the meteorological index evaluation model is greater than the credibility threshold, the index label of the meteorological index evaluation model is generated; A meteorological index-index label mapping table is established, and the corresponding relationship between the meteorological index evaluation model and the meteorological index is generated according to the meteorological index-index label mapping table.

5. The multi-source weather data integration and intelligent analysis system of claim 4, wherein, Based on the meteorological index evaluation model and the priority coefficient, the index evaluation value of each meteorological index of the current meteorological data is determined, including: Obtaining the meteorological index of the current meteorological data, and screening out the meteorological index evaluation model corresponding to the current meteorological index based on the meteorological index-index label mapping table; The index related data of the current preset period of the meteorological index is input into the corresponding meteorological index evaluation model to obtain the first index evaluation value of the current meteorological index; According to the first index evaluation value and the priority coefficient of the corresponding meteorological index, the index evaluation value G of the current meteorological index is generated; G=g1*K; Wherein, g1 is the first index evaluation value of the current meteorological index.

6. The multi-source weather data integration and intelligent analysis system of claim 5, wherein, According to the index evaluation value of each meteorological index of the same meteorological data, the prediction result of the corresponding meteorological data is generated, including: The index evaluation value of each meteorological index of the same meteorological data is processed by mean value to obtain the evaluation mean value of the corresponding meteorological data; The first preset evaluation mean interval, the second preset evaluation mean interval and the third preset evaluation mean interval are preset; When the evaluation mean of the meteorological data is in the first preset evaluation mean interval, the prediction result of the current meteorological data is false prediction; When the evaluation mean of the meteorological data is in the second preset evaluation mean interval, the prediction result of the current meteorological data is deviation prediction; When the evaluation mean of the meteorological data is in the third preset evaluation mean interval, the prediction result of the current meteorological data is accurate prediction.

7. The multi-source weather data integration and intelligent analysis system of claim 6, wherein, Before generating the prediction result corresponding to the meteorological data, the method further comprises: Obtaining the configuration requirement of the user end, and filtering out specific meteorological data according to the configuration requirement and setting the configuration parameter of the index configuration component of the specific meteorological data; Generating the index evaluation value of the specific meteorological data under the corresponding configuration parameter according to the index configuration component, and performing visual display according to the preset display form.

Images