Multi-source meteorological data integration and intelligent analysis system

Abstract

The invention discloses a multi-source meteorological data integration and intelligent analysis system, and the system comprises an establishment module which is used for establishing a plurality of meteorological indexes of multi-source meteorological data, obtaining the historical information of the meteorological indexes, and setting the priority coefficients of the corresponding meteorological indexes according to the historical information; the generation module is used for generating a meteorological index evaluation model according to the meteorological indexes and establishing a corresponding relation between the meteorological indexes and the meteorological index evaluation model; the evaluation module is used for calculating meteorological indexes of the meteorological data in the preset time period, and determining an index evaluation value of each meteorological index of the current meteorological data based on a meteorological index evaluation model and the priority coefficient; and the visualization module is used for generating a forecast result of the corresponding meteorological data according to the index evaluation value of each meteorological index of the same meteorological data and performing visual display. By improving the evaluation accuracy of meteorological data forecast, meteorological data meeting user requirements can be quickly provided, and the accuracy and timeliness of new energy market transaction decision making are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of meteorological data analysis, and in particular to a multi-source meteorological data integration and intelligent analysis system. Background Art

[0002] The prediction of new energy power is to input the forecast conclusions of meteorological elements into the power generation model, and then calculate how many kilowatt-hours of electricity the wind and photovoltaic fields can generate, and then form a production scheduling plan based on this, so as to make good operation arrangements and power transmission. Therefore, the meteorological data forecast results are one of the important factors affecting the accuracy of new energy power prediction.

[0003] In the existing technology, most of them use a single meteorological calculation indicator to evaluate the accuracy of meteorological data. The above method is simple and less comprehensive, which reduces the evaluation accuracy of meteorological data forecasts. In addition, the amount of meteorological data processed is large, and it is impossible to quickly provide meteorological data that meets user needs, which reduces the accuracy and timeliness of new energy market transaction decisions. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a multi-source meteorological data integration and intelligent analysis system, which integrates and processes multi-source meteorological data and establishes multiple meteorological indicators, constructs multiple meteorological indicator evaluation models, and obtains the indicator evaluation value of each meteorological indicator according to the meteorological indicator evaluation model and the priority coefficient. The current meteorological data forecast result is obtained according to the indicator evaluation values ​​of multiple meteorological indicators of the same meteorological data and is visualized, thereby improving the evaluation accuracy of the meteorological data forecast, and can quickly provide meteorological data that meets user needs, thereby improving the accuracy and timeliness of new energy market transaction decisions.

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

[0006] Establish a module for establishing several meteorological indicators of multi-source meteorological data, obtaining historical information of meteorological indicators, and setting priority coefficients of corresponding meteorological indicators according to the historical information;

[0007] A generation module is used to generate a meteorological index evaluation model according to the meteorological index, and establish a corresponding relationship between the meteorological index and the meteorological index evaluation model;

[0008] An evaluation module, used to calculate the meteorological index of the meteorological data of a preset period, and determine the index evaluation value of each meteorological index of the current meteorological data based on the meteorological index evaluation model and the priority coefficient;

[0009] The visualization module is used to generate the forecast results of the corresponding meteorological data according to the indicator evaluation value 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 root mean square error, completeness rate, efficiency, correlation and MAE of multi-source meteorological data forecasts.

[0011] In some implementations of the present invention, setting a priority coefficient of a corresponding meteorological indicator according to historical information includes:

[0012] Acquire historical information of meteorological indicators, wherein the historical information includes generation identification, correlation identification and application identification;

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

[0014] According to multiple completion marks of meteorological indicators, historical meteorological indicator change curves of meteorological indicators corresponding to the multiple completion marks are obtained, and the forecast rate average of the historical forecast results of the historical meteorological indicator change curve is generated according to the multiple historical meteorological indicator change curves and the corresponding historical forecast results, and the number of abnormal historical meteorological indicator change curves is screened out according to the forecast success rate threshold, and the forecast stability of the current meteorological indicator is generated;

[0015] Determine a first application frequency and a second application frequency according to an application identifier of the meteorological indicator, obtain a first impact weight of the number of times the corresponding meteorological indicator is used alone according to the first application frequency, obtain a second impact weight of the number of times the corresponding meteorological indicator is used together according to the second application frequency, and generate a comprehensive impact weight of the corresponding meteorological indicator according to the first impact weight and the second impact weight;

[0016] The priority coefficient of the corresponding meteorological indicator is generated according to the generation time, forecast stability and comprehensive impact weight of the meteorological indicator.

[0017] In some embodiments of the present invention, the priority coefficient of the corresponding meteorological indicator is generated according to the generation time, forecast stability and comprehensive impact weight of the meteorological indicator, including:

[0018] The calculation formula of the priority coefficient is:

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

[0020] Wherein, 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 impact weight conversion coefficient, e3 is the weight coefficient corresponding to the comprehensive impact weight, (u1*r1+u2*r2) / n is the comprehensive impact weight, u1 is the first impact weight, r1 is the weight coefficient of the first impact weight, u2 is the second impact weight, r2 is the weight coefficient of the second impact weight, and n is the combined number of single use and combined use;

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

[0022] Presetting a first preset priority coefficient threshold and a second preset priority coefficient threshold;

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

[0024] When the preferred coefficient is within a first preset priority coefficient threshold and a second preset priority coefficient threshold, setting a secondary tag to the corresponding weather indicator;

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

[0026] In some embodiments of the present invention, generating a meteorological index evaluation model according to meteorological indexes includes:

[0027] Obtain the indicator-related data and historical indicator evaluation values ​​of each meteorological indicator in the historical period;

[0028] Generate training set data and test set data based on indicator-related data and historical indicator evaluation values;

[0029] Generate a meteorological index evaluation model based on the training set data, and generate the credibility of the meteorological index evaluation model 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 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;

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

[0033] A meteorological indicator-index label mapping table is established, and the corresponding relationship between the meteorological indicator evaluation model and the meteorological indicator is generated according to the meteorological indicator-index label mapping table.

[0034] In some embodiments of the present invention, based on the meteorological index evaluation model and the priority coefficient, determining the index evaluation value of each meteorological index of the current meteorological data includes:

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

[0036] Inputting the index-related data of the current preset time period of the meteorological index into the corresponding meteorological index evaluation model to obtain the first index evaluation value of the current meteorological index;

[0037] Generate an index evaluation value G of the current meteorological index according to the first index evaluation value and the priority coefficient of the corresponding meteorological index;

[0038] G = g1*K;

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

[0040] In some embodiments of the present invention, generating a forecast result of corresponding meteorological data according to the index evaluation value of each meteorological index of the same meteorological data includes:

[0041] The index evaluation value of each meteorological index of the same meteorological data is averaged to obtain the evaluation mean of the corresponding meteorological data;

[0042] Presetting a first preset evaluation mean interval, a second preset evaluation mean interval and a third preset evaluation mean interval;

[0043] When the evaluation mean of the meteorological data is within the first preset evaluation mean interval, the forecast result of the current meteorological data is an erroneous prediction;

[0044] When the evaluation mean of the meteorological data is within the second preset evaluation mean interval, the forecast result of the current meteorological data is a deviation forecast;

[0045] When the evaluation mean of the meteorological data is within the third preset evaluation mean interval, 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 of the user end, filter out specific meteorological data according to the configuration requirements, and set the configuration parameters of the indicator configuration component of the specific meteorological data;

[0048] Generate indicator evaluation values ​​under corresponding configuration parameters of specific meteorological data according to indicator configuration components, and display them visually in the preset display format.

[0049] Compared with the prior art, the multi-source meteorological data integration and intelligent analysis system provided by the present invention has the following beneficial effects:

[0050] By integrating and processing multi-source meteorological data and establishing multiple meteorological indicators, multiple meteorological indicator evaluation models are constructed. According to the meteorological indicator evaluation model and the priority coefficient, the indicator evaluation value of each meteorological indicator is obtained. According to the indicator evaluation values ​​of multiple meteorological indicators of the same meteorological data, the current meteorological data forecast results are obtained and visualized, thereby improving the evaluation accuracy of meteorological data forecasts, quickly providing meteorological data that meets user needs, and improving the accuracy and timeliness of new energy market transaction decisions. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0052] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0053] In the description of the present invention, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship are based on the orientation or position relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

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

[0055] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

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

[0057] Establish a module for establishing several meteorological indicators of multi-source meteorological data, obtaining historical information of meteorological indicators, and setting priority coefficients of corresponding meteorological indicators according to the historical information;

[0058] A generation module is used to generate a meteorological index evaluation model according to the meteorological index, and establish a corresponding relationship between the meteorological index and the meteorological index evaluation model;

[0059] An evaluation module, used to calculate the meteorological index of the meteorological data of a preset period, and determine the index evaluation value of each meteorological index of the current meteorological data based on the meteorological index evaluation model and the priority coefficient;

[0060] The visualization module is used to generate the forecast results of the corresponding meteorological data according to the indicator evaluation value 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 resolutions and spatial resolutions, and can support meteorological data in multiple formats and multiple access methods, including temporal resolution, spatial resolution, meteorological file format, access method, etc., to realize integrated processing of multi-source meteorological data.

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

[0063] In this embodiment, a data foundation is laid for calculating the indicator evaluation value of the meteorological indicator by calculating the priority coefficient of each meteorological indicator, and important indicators of multiple meteorological indicators are screened out according to the priority coefficients. The important indicators are calculated and cached in advance. When the user calls the corresponding meteorological indicator, the cached content can be directly displayed, 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 rate, efficiency, correlation and MAE (mean absolute error) of multi-source meteorological data forecasts.

[0065] In some embodiments of the present invention, setting the priority coefficient of the corresponding meteorological indicator according to the historical information includes:

[0066] Acquire historical information of meteorological indicators, wherein the historical information includes generation identification, correlation identification and application identification;

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

[0068] According to multiple completion marks of meteorological indicators, historical meteorological indicator change curves of meteorological indicators corresponding to the multiple completion marks are obtained, and the forecast rate average of the historical forecast results of the historical meteorological indicator change curve is generated according to the multiple historical meteorological indicator change curves and the corresponding historical forecast results, and the number of abnormal historical meteorological indicator change curves is screened out according to the forecast success rate threshold, and the forecast stability of the current meteorological indicator is generated;

[0069] Determine a first application frequency and a second application frequency according to an application identifier of the meteorological indicator, obtain a first impact weight of the number of times the corresponding meteorological indicator is used alone according to the first application frequency, obtain a second impact weight of the number of times the corresponding meteorological indicator is used together according to the second application frequency, and generate a comprehensive impact weight of the corresponding meteorological indicator according to the first impact weight and the second impact weight;

[0070] The priority coefficient of the corresponding meteorological indicator is generated according to the generation time, forecast stability and comprehensive impact weight of the meteorological indicator.

[0071] In this embodiment, the generation identifier of each meteorological indicator is different. The generation identifier refers to each meteorological indicator calculated using at least one generation formula. The completion identifier refers to the identifier when the meteorological indicator calculation is completed. The corresponding historical meteorological indicator change curve is generated according to the corresponding meteorological indicator data when the calculation is completed. According to 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 refers to a curve whose historical forecast results are greater than the average forecast rate. 1-the number of abnormal historical meteorological indicator change curves / the number of normal historical meteorological indicator change curves = forecast stability.

[0072] In this embodiment, the first application frequency refers to the influence of the corresponding meteorological indicator on the forecast result 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 influence of the corresponding meteorological indicator on the forecast result 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, the priority coefficient of the meteorological indicator is generated according to the generation time, forecast stability and comprehensive impact weight of the meteorological indicator, laying a data foundation for the subsequent calculation of the indicator evaluation value. In order to facilitate the user's timely scheduling of the corresponding meteorological indicators, the meteorological indicators with high priority coefficients are cached first, thereby ensuring the user experience and improving the efficient use of meteorological data.

[0074] In some embodiments of the present invention, the priority coefficient of the corresponding meteorological indicator is generated according to the generation time, forecast stability and comprehensive impact weight of the meteorological indicator, including:

[0075] The calculation formula of the priority coefficient is:

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

[0077] Wherein, 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 impact weight conversion coefficient, e3 is the weight coefficient corresponding to the comprehensive impact weight, (u1*r1+u2*r2) / n is the comprehensive impact weight, u1 is the first impact weight, r1 is the weight coefficient of the first impact weight, u2 is the second impact weight, r2 is the weight coefficient of the second impact weight, and n is the combined number of single use and combined use;

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

[0079] Presetting a first preset priority coefficient threshold and a second preset priority coefficient threshold;

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

[0081] When the preferred coefficient is within a first preset priority coefficient threshold and a second preset priority coefficient threshold, setting a secondary tag to the corresponding weather indicator;

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

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

[0084] In some embodiments of the present invention, generating a meteorological index evaluation model according to meteorological indexes includes:

[0085] Obtain the indicator-related data and historical indicator evaluation values ​​of each meteorological indicator in the historical period;

[0086] Generate training set data and test set data based on indicator-related data and historical indicator evaluation values;

[0087] Generate a meteorological index evaluation model based on the training set data, and generate the credibility of the meteorological index evaluation model 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 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;

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

[0091] A meteorological indicator-index label mapping table is established, and the corresponding relationship between the meteorological indicator evaluation model and the meteorological indicator is generated according to the meteorological indicator-index label mapping table.

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

[0093] In some embodiments of the present invention, based on the meteorological index evaluation model and the priority coefficient, determining the index evaluation value of each meteorological index of the current meteorological data includes:

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

[0095] Inputting the index-related data of the current preset time period of the meteorological index into the corresponding meteorological index evaluation model to obtain the first index evaluation value of the current meteorological index;

[0096] Generate an index evaluation value G of the current meteorological index according to the first index evaluation value and the priority coefficient of the corresponding meteorological index;

[0097] G = g1*K;

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

[0099] In some embodiments of the present invention, generating a forecast result of corresponding meteorological data according to the index evaluation value of each meteorological index of the same meteorological data includes:

[0100] The index evaluation value of each meteorological index of the same meteorological data is averaged to obtain the evaluation mean of the corresponding meteorological data;

[0101] Presetting a first preset evaluation mean interval, a second preset evaluation mean interval and a third preset evaluation mean interval;

[0102] When the evaluation mean of the meteorological data is within the first preset evaluation mean interval, the forecast result of the current meteorological data is an erroneous prediction;

[0103] When the evaluation mean of the meteorological data is within the second preset evaluation mean interval, the forecast result of the current meteorological data is a deviation forecast;

[0104] When the evaluation mean of the meteorological data is within the third preset evaluation mean interval, 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 the historical forecast results. The more accurate the forecast result is, the larger the evaluation mean should be. An erroneous prediction means that the forecast result accounts for 30% of the correct result, a deviation prediction means that the forecast result accounts for 60% of the correct result, and an accurate prediction means that the forecast result accounts for 90% or more of the correct result.

[0106] In this embodiment, an evaluation mean is obtained based on the indicator evaluation values ​​of multiple meteorological indicators, thereby improving the evaluation accuracy of meteorological data forecasts, thereby accurately obtaining the meteorological data forecast performance based on meteorological indicators, screening out meteorological data with high forecast accuracy, and improving the accuracy of new energy power prediction, thereby improving the accuracy and timeliness of new energy market transaction decisions.

[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 of the user end, filter out specific meteorological data according to the configuration requirements, and set the configuration parameters of the indicator configuration component of the specific meteorological data;

[0109] Generate indicator evaluation values ​​under corresponding configuration parameters of specific meteorological data according to indicator configuration components, and display them visually in the preset display format.

[0110] In this embodiment, the configuration requirements of the user end include date range and specific meteorological data, as well as specific scenario requirements of specific meteorological data or custom meteorological indicators. The specific scenario requirements include time periods of heavy and heavy winds, peak and trough periods of wind speed, wind speed, etc. The indicator configuration component refers to custom configuration settings that support various indicators, such as forecast day settings, time period settings, and threshold settings. The user end can select and configure the required meteorological indicators on the visual interface, and provide an indicator configuration interface to allow users to choose to add, delete, sort, and combine indicators from the list of available meteorological indicators. Users can define the presentation format of meteorological indicators, including line charts, bar charts, pie charts, etc., and configure attributes such as color, label, and unit.

[0111] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention. These improvements and substitutions should also be regarded as the scope of protection of the present invention.

Claims

1. A multi-source meteorological data integration and intelligent analysis system, characterized in that: include: Establish a module for establishing several meteorological indicators of multi-source meteorological data, obtaining historical information of meteorological indicators, and setting priority coefficients of corresponding meteorological indicators according to the historical information; A generation module is used to generate a meteorological index evaluation model according to the meteorological index, and establish a corresponding relationship between the meteorological index and the meteorological index evaluation model; An evaluation module, used to calculate the meteorological index of the meteorological data of a preset period, and determine the index evaluation value of each meteorological index of the current meteorological data based on the meteorological index evaluation model and the priority coefficient; The visualization module is used to generate the forecast results of the corresponding meteorological data according to the indicator evaluation value of each meteorological indicator of the same meteorological data, and to visualize the forecast results.

2. The multi-source meteorological data integration and intelligent analysis system according to claim 1, characterized in that: The meteorological indicators include root mean square error, completeness rate, efficiency, correlation and MAE of multi-source meteorological data forecast.

3. The multi-source meteorological data integration and intelligent analysis system according to claim 2, characterized in that: The priority coefficients of corresponding meteorological indicators are set according to historical information, including: Acquire historical information of meteorological indicators, wherein the historical information includes generation identification, correlation identification and application identification; According to the generation identifier of the meteorological indicator, the initial acquisition form and the final generation form of the current meteorological indicator are determined; according to the initial acquisition form and the final generation form, relevant formulas are retrieved from the identifier-formula database, and the relevant formulas are simulated to obtain the calculation time of each relevant formula; and according to the calculation sequence and calculation time of the relevant formulas, the generation framework of the corresponding meteorological indicator is obtained, and the generation time of the corresponding meteorological indicator is set according to the route length of the generation framework; According to multiple completion marks of meteorological indicators, historical meteorological indicator change curves of meteorological indicators corresponding to the multiple completion marks are obtained, and the forecast rate average of the historical forecast results of the historical meteorological indicator change curve is generated according to the multiple historical meteorological indicator change curves and the corresponding historical forecast results, and the number of abnormal historical meteorological indicator change curves is screened out according to the forecast success rate threshold, and the forecast stability of the current meteorological indicator is generated; Determine a first application frequency and a second application frequency according to an application identifier of the meteorological indicator, obtain a first impact weight of the number of times the corresponding meteorological indicator is used alone according to the first application frequency, obtain a second impact weight of the number of times the corresponding meteorological indicator is used together according to the second application frequency, and generate a comprehensive impact weight of the corresponding meteorological indicator according to the first impact weight and the second impact weight; The priority coefficient of the corresponding meteorological indicator is generated according to the generation time, forecast stability and comprehensive impact weight of the meteorological indicator.

4. The multi-source meteorological data integration and intelligent analysis system according to claim 3, characterized in that: The priority coefficients of the corresponding meteorological indicators are generated according to the generation time, forecast stability and comprehensive impact weight of the meteorological indicators, including: The calculation formula of the priority coefficient is: K=a1*e1*t+a2*e2*W+a3*e3*(u1*r1+u2*r2) / n; Wherein, 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 impact weight conversion coefficient, e3 is the weight coefficient corresponding to the comprehensive impact weight, (u1*r1+u2*r2) / n is the comprehensive impact weight, u1 is the first impact weight, r1 is the weight coefficient of the first impact weight, u2 is the second impact weight, r2 is the weight coefficient of the second impact weight, and n is the combined number of single use and combined use; According to the relationship between the priority coefficients of multiple meteorological indicators of the same meteorological data and the preset priority coefficient threshold, the label information of the corresponding meteorological indicator is set; Presetting a first preset priority coefficient threshold and a second preset priority coefficient threshold; When the preferred coefficient is at a first preset priority coefficient threshold, an ignore tag is set for the corresponding meteorological indicator; When the preferred coefficient is within a first preset priority coefficient threshold and a second preset priority coefficient threshold, setting a secondary tag to the corresponding weather indicator; When the preferred coefficient is greater than the second preset priority coefficient threshold, an important label is set for the corresponding meteorological indicator and cache processing is performed.

5. The multi-source meteorological data integration and intelligent analysis system according to claim 4, characterized in that: Generate a meteorological indicator evaluation model based on meteorological indicators, including: Obtain the indicator-related data and historical indicator evaluation values ​​of each meteorological indicator in the historical period; Generate training set data and test set data based on indicator-related data and historical indicator evaluation values; Generate a meteorological index evaluation model based on the training set data, and generate the credibility of the meteorological index evaluation model based on the test set data; Preset credibility threshold; 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, an index label of the meteorological index evaluation model is generated; A meteorological indicator-index label mapping table is established, and the corresponding relationship between the meteorological indicator evaluation model and the meteorological indicator is generated according to the meteorological indicator-index label mapping table.

6. The multi-source meteorological data integration and intelligent analysis system according to claim 5, characterized in that: 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: Obtain the meteorological index of the current meteorological data, and filter out the meteorological index evaluation model corresponding to the current meteorological index based on the meteorological index-index label mapping table; Inputting the index-related data of the current preset time period of the meteorological index into the corresponding meteorological index evaluation model to obtain the first index evaluation value of the current meteorological index; Generate an index evaluation value G of the current meteorological index according to the first index evaluation value and the priority coefficient of the corresponding meteorological index; G = g1*K; Among them, g1 is the first indicator evaluation value of the current meteorological indicator.

7. The multi-source meteorological data integration and intelligent analysis system according to claim 6, characterized in that: According to the index evaluation value of each meteorological index of the same meteorological data, the forecast results of the corresponding meteorological data are generated, including: The index evaluation value of each meteorological index of the same meteorological data is averaged to obtain the evaluation mean of the corresponding meteorological data; Presetting a first preset evaluation mean interval, a second preset evaluation mean interval and a third preset evaluation mean interval; When the evaluation mean of the meteorological data is within the first preset evaluation mean interval, the forecast result of the current meteorological data is an erroneous prediction; When the evaluation mean of the meteorological data is within the second preset evaluation mean interval, the forecast result of the current meteorological data is a deviation forecast; When the evaluation mean of the meteorological data is within the third preset evaluation mean interval, the forecast result of the current meteorological data is an accurate prediction.

8. The multi-source meteorological data integration and intelligent analysis system according to claim 7, characterized in that: Before generating the forecast results corresponding to the meteorological data, it also includes: Obtain the configuration requirements of the user end, and according to the configuration requirements, filter out specific meteorological data and set the configuration parameters of the indicator configuration component of the specific meteorological data; Generate indicator evaluation values ​​under corresponding configuration parameters of specific meteorological data according to indicator configuration components, and display them visually in the preset display format.

Images