Weather forecast data processing method and device based on dynamic weight

Abstract

The invention discloses a weather forecast data processing method and device based on dynamic weight, and the method comprises the steps: obtaining a forecast data information set and analysis data information; the forecast data information set comprises N pieces of forecast data information; n is a positive integer; performing fusion processing on the forecast data information set and the analysis data information to obtain target weight information; according to the target weight information, processing the forecast data information set of the next time period to obtain target forecast information; the next time period forecast data information set comprises N pieces of next time period forecast data information. Therefore, according to the invention, stable and comprehensively integrated objective forecast result information is obtained through the multi-mode forecast data information, and the weather forecast service capability is improved.

Description

Technical Field

[0001] The present invention relates to the application field of numerical weather forecast, and in particular to a weather forecast data processing method and device based on dynamic weight. Background Art

[0002] The meteorological environment is complex and changeable, and various severe weather phenomena often have disastrous effects on production and life. Therefore, it is necessary to continuously improve the methods and means of operational meteorological forecasting and rapidly improve the ability of refined meteorological forecasting services.

[0003] With the rapid development of numerical weather forecasting technology, multiple numerical models of different scales have provided new opportunities for improving forecasting capabilities. The types of numerical forecast data from different models are increasing day by day, but because the "singleness" and "uncertainty" shown by a single model in the forecast process affect the accuracy of the model forecast, each model has its own advantages and disadvantages under different geographical conditions. Limited by the model level, assimilation ability, and computing conditions, it is difficult to improve the accuracy of numerical forecasts. How to use multi-model forecast data information to obtain stable, comprehensive, integrated, and objective forecast result information is an effective way to improve numerical weather forecast products and an important application for the interpretation of numerical forecast products. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a weather forecast data processing method and device based on dynamic weights, which can obtain stable, comprehensive and integrated objective forecast result information through multi-mode forecast data information, which is conducive to improving the meteorological forecast service capabilities.

[0005] In order to solve the above technical problems, a first aspect of an embodiment of the present invention discloses a weather forecast data processing method based on dynamic weights, the method comprising:

[0006] S1, obtaining a forecast data information set and analysis data information; the forecast data information set includes N forecast data information; N is a positive integer;

[0007] S2, fusing the forecast data information set and the analysis data information to obtain target weight information;

[0008] S3, processing the next period forecast data information set according to the target weight information to obtain target forecast information; the next period forecast data information set includes N next period forecast data information.

[0009] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the fusing the forecast data information set and the analysis data information to obtain the target weight information includes:

[0010] S21, preprocessing the forecast data information set and the analysis data information to obtain a preprocessed forecast data information set and preprocessed analysis data information; the preprocessed forecast data information set includes the N preprocessed forecast data information;

[0011] S22, processing the pre-processed forecast data information set and the pre-processed analysis data information to obtain a forecast data information set to be processed and analysis data information to be processed; the forecast data information set to be processed includes the N forecast data information to be processed; the forecast data information to be processed includes M forecast data information at time points to be processed; the analysis data information to be processed includes M analysis data information at time points to be processed; and M is a positive integer;

[0012] S23, performing calculation processing on the forecast data information set to be processed and the analysis data information to be processed to obtain forecast threshold information to be processed and a forecast average threshold;

[0013] S24, calculating and processing the forecast threshold information to be processed and the forecast average threshold to obtain target weight information; the target weight information includes the N target forecast weight values.

[0014] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the preprocessing of the forecast data information set and the analysis data information to obtain the preprocessed forecast data information set and the preprocessed analysis data information includes:

[0015] S211, performing data cleaning processing on the forecast data information set and the analysis data information to obtain a cleaned forecast data information set and cleaned analysis data information;

[0016] S212, performing bilinear interpolation processing on the cleaned forecast data information set to obtain an interpolated forecast data information set;

[0017] S213, performing boundary check and category consistency check processing on the interpolated forecast data information set and the cleaned analysis data information respectively to obtain a preprocessed forecast data information set and preprocessed analysis data information.

[0018] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the processing of the preprocessed forecast data information set and the preprocessed analysis data information to obtain the forecast data information set to be processed and the analysis data information to be processed includes:

[0019] S221, obtaining time window range information and time interval value;

[0020] S222, performing a first judgment process on the preprocessing forecast data information set and the preprocessing analysis data information according to the time window range information to obtain a first preprocessing forecast data information set and a first preprocessing analysis data information;

[0021] S223: Perform a second judgment process on the first preprocessed forecast data information set and the first preprocessed analysis data information according to the time window range information and the time interval value to obtain a forecast data information set to be processed and an analysis data information to be processed.

[0022] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the calculating and processing the forecast data information set to be processed and the analysis data information to be processed to obtain the forecast threshold information to be processed and the forecast average threshold value includes:

[0023] S231, using a weather forecast first weight calculation model, calculating and processing the forecast data information set to be processed and the analysis data information to be processed to obtain forecast threshold information to be processed;

[0024] Wherein, the weather forecast first weight calculation model is:

[0025]

[0026] Where, T is the forecast threshold information to be processed, T i is the i-th forecast threshold in the forecast threshold information to be processed, F i,j,k is the kth forecast data value in the jth forecast data information at the time point to be processed in the i-th forecast data information to be processed in the forecast data information set to be processed, O j,k is the kth analysis data value in the jth analysis data information to be processed at the time point to be processed in the analysis data information to be processed, L i,j is the number of forecast data values ​​in the forecast data information at the jth time point to be processed in the ith forecast data information to be processed in the forecast data information set to be processed, M is the number of forecast data information at the time point to be processed in the forecast data information to be processed, and N is the number of forecast data information to be processed in the forecast data information set to be processed;

[0027] S232, using a second weight calculation model for weather forecast, calculating and processing the forecast threshold information to be processed to obtain a forecast average threshold;

[0028] Wherein, the second weight calculation model of weather forecast is:

[0029]

[0030] In the formula, is the forecast average threshold.

[0031] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the calculating and processing the forecast threshold information to be processed and the forecast average threshold to obtain the target weight information includes:

[0032] S241, determining whether any of the forecast thresholds in the forecast threshold information to be processed is greater than the forecast average threshold, and obtaining a threshold determination result;

[0033] When the threshold judgment result is yes, determining the preset forecast weight value to be the target forecast weight value corresponding to the forecast threshold;

[0034] When the threshold judgment result is no, executing S242;

[0035] S242, using a third weather forecast weight calculation model, calculating and processing the forecast threshold, the forecast threshold information to be processed, and the forecast average threshold, to obtain a target forecast weight value corresponding to the forecast threshold;

[0036] Wherein, the third weight calculation model of weather forecast is:

[0037]

[0038] Wherein, ω is the target forecast weight value corresponding to the forecast threshold, D is the forecast threshold, T is the forecast threshold information to be processed, and T b is the bth forecast threshold value in the forecast threshold information to be processed that is less than or equal to the average threshold value, and V is the number of the forecast threshold values ​​in the forecast threshold information to be processed that are less than or equal to the average threshold value.

[0039] As an optional implementation manner, in the first aspect of the embodiment of the present invention, the processing of the next time period forecast data information set according to the target weight information to obtain the target forecast information includes:

[0040] S31, preprocessing the next period forecast data information set to obtain a next period preprocessed forecast data information set; the next period preprocessed forecast data information set includes N next period preprocessed forecast data information;

[0041] S32, using the fourth weight calculation model of weather forecast, calculating and processing the target weight information and the next period pre-processed forecast data information set to obtain target forecast information;

[0042] Wherein, the fourth weight calculation model of weather forecast is:

[0043]

[0044] Where, F′ is the target forecast information, F″ i1 is the i1th next period pre-processed forecast data information in the next period pre-processed forecast data information set, ω i1 is the i1th target forecast weight value in the target weight information, and N is the number of the next time period pre-processed forecast data information in the next time period pre-processed forecast data information set.

[0045] A second aspect of an embodiment of the present invention discloses a weather forecast data processing device, the device comprising:

[0046] An acquisition module, used for acquiring a forecast data information set and analysis data information; the forecast data information set includes N forecast data information; N is a positive integer;

[0047] A first calculation module is used to fuse the forecast data information set and the analysis data information to obtain target weight information;

[0048] The second calculation module is used to process the next period forecast data information set according to the target weight information to obtain target forecast information; the next period forecast data information set includes N next period forecast data information.

[0049] A third aspect of an embodiment of the present invention discloses another weather forecast data processing device based on dynamic weights, the device comprising:

[0050] processor;

[0051] a memory coupled to the processor and storing executable program code;

[0052] The processor calls the executable program code stored in the memory to execute part or all of the steps of the weather forecast data processing method based on dynamic weights disclosed in the first aspect of the embodiment of the present invention.

[0053] The fourth aspect of an embodiment of the present invention discloses a computer-readable storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute part or all of the steps of the weather forecast data processing method based on dynamic weights disclosed in the first aspect of the embodiment of the present invention.

[0054] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0055] In an embodiment of the present invention, a forecast data information set and an analysis data information are obtained; the forecast data information set includes N forecast data information; N is a positive integer; the forecast data information set and the analysis data information are fused to obtain target weight information; according to the target weight information, the next time period forecast data information set is processed to obtain target forecast information; the next time period forecast data information set includes N next time period forecast data information. It can be seen that the present application obtains stable, comprehensive and integrated objective forecast result information through multi-mode forecast data information, which is conducive to improving the meteorological forecast service capability. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0057] Figure 1 A schematic flow chart of a weather forecast data processing method based on dynamic weights disclosed in an embodiment of the present invention;

[0058] Figure 2 A schematic diagram of the structure of a weather forecast data processing device based on dynamic weights disclosed in an embodiment of the present invention;

[0059] Figure 3 The present invention is a schematic diagram of the structure of another weather forecast data processing device based on dynamic weights disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or equipment that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or equipment.

[0062] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0063] The present invention discloses a weather forecast data processing method and device based on dynamic weights, which obtains stable, comprehensive and integrated objective forecast result information through multi-mode forecast data information, and is beneficial to improving the weather forecast service capability.

[0064] Embodiment 1

[0065] See also Figure 1 , Figure 1 : is a flow chart of a weather forecast data processing method based on dynamic weights disclosed in an embodiment of the present invention. Figure 1 The described weather forecast data processing method based on dynamic weights is applied to a weather forecast data processing device based on dynamic weights, such as a local server or cloud server for optimizing and managing weather forecast data processing based on dynamic weights, and the embodiments of the present invention do not limit this. Figure 1 As shown, the weather forecast data processing method based on dynamic weights may include the following operations:

[0066] S1, obtaining a forecast data information set and analysis data information; the forecast data information set includes N forecast data information; N is a positive integer;

[0067] It should be noted that the N forecast data information are GRAPE forecast data, ECMWF forecast data, GFS forecast data, and WRF forecast data; the analysis data information is EC reanalysis data.

[0068] Exemplarily, N is 4.

[0069] It should be noted that the forecast data information includes several forecast data values, and the analysis data information includes several analysis data values.

[0070] S2, fusing the forecast data information set and the analysis data information to obtain target weight information;

[0071] S3, processing the next period forecast data information set according to the target weight information to obtain target forecast information; the next period forecast data information set includes N next period forecast data information.

[0072] It can be seen that the implementation of the weather forecast data processing method based on dynamic weights described in the embodiment of the present invention can obtain stable, comprehensive and integrated objective forecast result information through multi-mode forecast data information, which is beneficial to improving the meteorological forecast service capabilities.

[0073] In an optional embodiment, the step of fusing the forecast data information set and the analysis data information to obtain target weight information includes:

[0074] S21, preprocessing the forecast data information set and the analysis data information to obtain a preprocessed forecast data information set and preprocessed analysis data information; the preprocessed forecast data information set includes the N preprocessed forecast data information;

[0075] S22, processing the pre-processed forecast data information set and the pre-processed analysis data information to obtain a forecast data information set to be processed and analysis data information to be processed; the forecast data information set to be processed includes the N forecast data information to be processed; the forecast data information to be processed includes M forecast data information at time points to be processed; the analysis data information to be processed includes M analysis data information at time points to be processed; and M is a positive integer;

[0076] S23, performing calculation processing on the forecast data information set to be processed and the analysis data information to be processed to obtain forecast threshold information to be processed and a forecast average threshold;

[0077] S24, calculating and processing the forecast threshold information to be processed and the forecast average threshold to obtain target weight information; the target weight information includes the N target forecast weight values.

[0078] It can be seen that the implementation of the weather forecast data processing method based on dynamic weights described in the embodiment of the present invention can obtain stable, comprehensive and integrated objective forecast result information through multi-mode forecast data information, which is beneficial to improving the meteorological forecast service capabilities.

[0079] In another optional embodiment, the preprocessing of the forecast data information set and the analysis data information to obtain the preprocessed forecast data information set and the preprocessed analysis data information includes:

[0080] S211, performing data cleaning processing on the forecast data information set and the analysis data information to obtain a cleaned forecast data information set and cleaned analysis data information;

[0081] It should be noted that the above data cleaning process is to extract forecast elements from the forecast data information set and the analysis data information, such as 2m air temperature, 2m humidity, 10m meridian wind, 10m zonal wind and other data information on the ground; the above data cleaning process for the forecast data information set and the analysis data information is to extract forecast elements from all the forecast data information and the analysis data information in the forecast data information set one by one. After extracting forecast elements from the forecast data information set, what is obtained is the cleaned forecast data information set, and after extracting forecast elements from the preprocessed analysis data information, what is obtained is the cleaned analysis data information. For the specific extraction method, tools such as Apache NiFi, Talend, and Informatica may be used, and other tools or algorithms may also be used, which are not limited in the embodiments of the present invention.

[0082] It should be noted that the above data cleaning can discover some new patterns that were not noticed before. For example, by analyzing the patterns of missing data, it can be found that weather data in certain regions or seasons is more likely to be missing, which may reflect the regularity or problems of weather monitoring, or even reveal the particularity of climate patterns, and integrate this information into the processing of weather forecasts.

[0083] S212, performing bilinear interpolation processing on the cleaned forecast data information set to obtain an interpolated forecast data information set;

[0084] It should be noted that the above-mentioned bilinear interpolation processing is obtained by calculation using a bilinear interpolation algorithm. Through bilinear interpolation, the low horizontal resolution data in the cleaning forecast data information set can be converted into high-resolution data consistent with the horizontal resolution of the cleaning analysis data information, so that subsequent processing can obtain more precise and accurate forecast results.

[0085] It should be noted that the above horizontal resolution refers to the horizontal distance between adjacent grid points in the grid data output by the weather forecast model. High resolution means that the distance between grid points is small, which can capture more detailed geographical and meteorological features; low resolution means that the distance between grid points is large, which can only capture coarse features. For example, high resolution is a 1 km resolution model, and the distance between each grid point is 1 km. Low resolution is a 25 km resolution model, and the distance between each grid point is 25 km.

[0086] S213, performing boundary check and category consistency check processing on the interpolated forecast data information set and the cleaned analysis data information respectively to obtain a preprocessed forecast data information set and preprocessed analysis data information;

[0087] It should be noted that the boundary check is to verify whether the forecast data values ​​and analysis data values ​​are within a reasonable and expected range. This is particularly important in the embodiments of the present invention, because data outside the reasonable range may indicate input errors, sensor failures, or other problems. Define the reasonable range of each variable based on meteorological common sense and historical data. For example, the temperature may be between -50°C and 50°C, and the wind speed may be between 0m / s and 100m / s. For outliers in boundary checks, a variety of methods can be used for processing, such as the nearest neighbor method, interpolation method, or machine learning prediction, etc., which are not limited in the embodiments of the present invention.

[0088] It should be noted that the category consistency check is to verify whether the classification attributes of the data meet expectations. The categories in the meteorological data may include weather types (sunny, rainy, etc.), wind directions (north wind, south wind, etc.), etc. The category consistency check can be performed by a simple matching method to match the interpolated forecast data information set and the categories in the cleaned analysis data information to see whether they are in the preset category information. If they are, they are filtered out, and if not, they are removed, thereby filtering out all the forecast data values ​​and analysis data values ​​in the preset category information. Specifically, the embodiments of the present invention are not limited.

[0089] It can be seen that the implementation of the weather forecast data processing method based on dynamic weights described in the embodiment of the present invention can obtain stable, comprehensive and integrated objective forecast result information through multi-mode forecast data information, which is beneficial to improving the meteorological forecast service capabilities.

[0090] In yet another optional embodiment, the processing of the preprocessed forecast data information set and the preprocessed analysis data information to obtain the forecast data information set to be processed and the analysis data information to be processed includes:

[0091] S221, obtaining time window range information and time interval value;

[0092] It should be noted that the time window information is in the range of 24 hours. For example, 00:00 on June 30, 2024 to 00:00 on July 1, 2024 is selected as the time window range information. The corresponding time window range start time point is 00:00 on June 30, 2024, and the time window range end time point is 00:00 on July 1, 2024.

[0093] Preferably, the time interval value is 6, in hours.

[0094] S222, performing a first judgment process on the preprocessing forecast data information set and the preprocessing analysis data information according to the time window range information to obtain a first preprocessing forecast data information set and a first preprocessing analysis data information;

[0095] It should be noted that the first judgment process is to judge whether the time information corresponding to the forecast data value or analysis data value in the preprocessed forecast data information set and the preprocessed analysis data information is between the start time point of the time window range and the end time point of the time window range. If so, the forecast data value or analysis data value is retained; if not, the forecast data value or analysis data value is discarded. For example, whether the time information corresponding to the forecast data value or analysis data value in the preprocessed forecast data information set and the preprocessed analysis data information is between 00:00 on June 30, 2024 and 00:00 on July 1, 2024.

[0096] S223: Perform a second judgment process on the first preprocessed forecast data information set and the first preprocessed analysis data information according to the time window range information and the time interval value to obtain a forecast data information set to be processed and an analysis data information to be processed.

[0097] It should be noted that, according to the time window range information and the time interval value, the second judgment processing of the first preprocessing forecast data information set and the first preprocessing analysis data information is to first generate M time points according to the time window range information and the time interval value, take the starting time point of the time window range information as the first time point, take the value obtained by subtracting the time interval value from the ending time point of the time window range information as the last time point, and take every other time interval value from the first time point to the last time point as a time point. Finally, for any time point among the M time points, the second judgment processing is used for judgment, when the time information corresponding to the forecast data value of any first preprocessing forecast data information in the first preprocessing forecast data information set is at the time point, it is added to the forecast data information of the time point to be processed corresponding to the time point, and when the time information corresponding to the analysis data value of the first preprocessing analysis data information is at the time point, it is added to the analysis data information of the time point to be processed corresponding to the time point. For example, 00:00 on June 30, 2024 to 00:00 on July 1, 2024 is the time window range information, the time interval value is 6, and the unit is hours, then the M time points obtained are 00:00 on June 30, 2024, 06:00 on June 30, 2024, 12:00 on June 30, 2024, and 18:00 on June 30, 2024. Then, for the time point 00:00 on June 30, 2024, for the first pre-processed forecast data information set, N forecast data information to be processed in the forecast data information set to be processed will be obtained, and each forecast data information to be processed will obtain 1 forecast data information to be processed at a time point corresponding to the time point; then, for the time point 00:00 on June 30, 2024, after the second judgment processing is performed on the first pre-processed analysis data information, 1 analysis data information to be processed at a time point in the analysis data information to be processed will be obtained.

[0098] It can be seen that the implementation of the weather forecast data processing method based on dynamic weights described in the embodiment of the present invention can obtain stable, comprehensive and integrated objective forecast result information through multi-mode forecast data information, which is beneficial to improving the meteorological forecast service capabilities.

[0099] In an optional embodiment, the calculating and processing the forecast data information set to be processed and the analysis data information to be processed to obtain forecast threshold information to be processed and a forecast average threshold value includes:

[0100] S231, using a weather forecast first weight calculation model, calculating and processing the forecast data information set to be processed and the analysis data information to be processed to obtain forecast threshold information to be processed;

[0101] Wherein, the weather forecast first weight calculation model is:

[0102]

[0103] Where, T is the forecast threshold information to be processed, T i is the i-th forecast threshold in the forecast threshold information to be processed, used to characterize the forecast threshold of the i-th forecast data information to be processed in the forecast data information set to be processed, F i,j,k is the kth forecast data value in the jth forecast data information at the time point to be processed in the i-th forecast data information to be processed in the forecast data information set to be processed, O j,k is the kth analysis data value in the jth analysis data information to be processed at the time point to be processed in the analysis data information to be processed, L i,j is the number of forecast data values ​​in the forecast data information at the jth time point to be processed in the ith forecast data information to be processed in the forecast data information set to be processed, M is the number of forecast data information at the time point to be processed in the forecast data information to be processed, and N is the number of forecast data information to be processed in the forecast data information set to be processed;

[0104] It should be noted that the forecast data value and the analysis data value in the embodiment of the present invention are data values ​​of the same attribute, and the specific attribute may be a temperature attribute, a meridional wind attribute, a zonal wind attribute, a humidity attribute, or other attributes, which are not limited in the embodiment of the present invention;

[0105] It should be noted that the implementation scheme in the embodiment of the present invention can perform dynamic weight calculation on one or more or all of the temperature attribute, meridional wind attribute, zonal wind attribute, and humidity attribute of the weather forecast at the same time, and the same attribute is used for each calculation. For example, when performing temperature calculation, the forecast data value and the analysis data value in the embodiment are both temperature values, and when performing humidity calculation, the forecast data value and the analysis data value in the embodiment are both humidity values.

[0106] S232, using a second weight calculation model for weather forecast, calculating and processing the forecast threshold information to be processed to obtain a forecast average threshold;

[0107] Wherein, the second weight calculation model of weather forecast is:

[0108]

[0109] In the formula, is the forecast average threshold.

[0110] It can be seen that the implementation of the weather forecast data processing method based on dynamic weights described in the embodiment of the present invention can obtain stable, comprehensive and integrated objective forecast result information through multi-mode forecast data information, which is beneficial to improving the meteorological forecast service capabilities.

[0111] In an optional embodiment, the calculating and processing the forecast threshold information to be processed and the forecast average threshold to obtain target weight information includes:

[0112] S241, determining whether any of the forecast thresholds in the forecast threshold information to be processed is greater than the forecast average threshold, and obtaining a threshold determination result;

[0113] When the threshold judgment result is yes, determining the preset forecast weight value to be the target forecast weight value corresponding to the forecast threshold;

[0114] When the threshold judgment result is no, executing S242;

[0115] Exemplarily, the preset forecast weight value is 0.

[0116] It should be noted that the above-mentioned judgment processing can simplify the calculation process and improve the operation efficiency. The preset forecast weight value and simple threshold judgment logic can avoid complex calculation processes. This design can significantly reduce the calculation delay and improve the response speed of the system. In addition, setting the preset forecast weight value to 0 can automatically filter low-quality or meaningless data, which is equivalent to adding a layer of "data noise reducer" to the system, which can reduce the interference of invalid data on the overall result during the forecast weight calculation process.

[0117] S242, using a third weather forecast weight calculation model, calculating and processing the forecast threshold, the forecast threshold information to be processed, and the forecast average threshold, to obtain a target forecast weight value corresponding to the forecast threshold;

[0118] Wherein, the third weight calculation model of weather forecast is:

[0119]

[0120] Wherein, ω is the target forecast weight value corresponding to the forecast threshold, D is the forecast threshold, T is the forecast threshold information to be processed, and T b is the bth forecast threshold value in the forecast threshold information to be processed that is less than or equal to the average threshold value, and V is the number of the forecast threshold values ​​in the forecast threshold information to be processed that are less than or equal to the average threshold value.

[0121] It should be noted that after executing the above S242, the updated target weight information can also be obtained through the following steps, and the updated target weight information can be determined as the target weight information; by performing fusion processing in the following ways, a more precise and accurate forecast result can be obtained.

[0122] S243, using the fifth weight calculation model of weather forecast, calculating and processing the forecast data information set to be processed and the analysis data information to be processed to obtain forecast data difference information;

[0123] Among them, the fifth weight calculation model of weather forecast is:

[0124]

[0125] Where, CY is the forecast data difference information, CY j1 is the j1th forecast data difference value in the forecast data difference information, E j1 is the j1th forecast data information to be processed in the forecast data information set to be processed, FX is the analysis data information to be processed, N is the number of forecast data information to be processed in the forecast data information set to be processed, represents the vector inner product, || represents the modulus of the vector, and δ1 is the first weight parameter;

[0126] It should be noted that the first weight parameter may be set by a user or obtained through historical data, and the embodiment of the present invention does not make any specific limitation thereto.

[0127] S244, using a sixth weight calculation model for weather forecast, processing the forecast data difference information to obtain first weight information;

[0128] Wherein, the sixth weight calculation model of weather forecast is:

[0129]

[0130] In the formula, DYQ is the first weight information, DYQ j1 is the j1th weight value in the first weight information, and δ2 is the second weight parameter;

[0131] It should be noted that the second weight parameter may be set by a user or obtained through historical data, and the embodiment of the present invention does not make any specific limitation thereto.

[0132] It should be noted that, through the fifth weight calculation model of weather forecast and the sixth weight calculation model of weather forecast, the difference between all the forecast data information to be processed in the forecast data information set and the analysis data information to be processed can be obtained, and the weights can be set according to the difference, so as to obtain more precise and accurate forecast results.

[0133] S245, constructing a matrix of the forecast data information set to be processed and the analysis data information to be processed to obtain a forecast data information matrix and an analysis data information matrix;

[0134] Wherein, the forecast data information matrix expression is:

[0135]

[0136] Where X is the forecast data information matrix, P i2,j2 is the j2th forecast data value in the i2th forecast data information to be processed in the forecast data information set to be processed;

[0137] Wherein, the analysis data information matrix is:

[0138]

[0139] Where, Y is the analysis data information matrix, Q i3 The i3th analysis data value in the analysis data information to be processed;

[0140] S246, using the seventh weight calculation model of weather forecast, calculating and processing the forecast data information matrix and the analysis data information matrix to obtain second weight information;

[0141] Among them, the seventh weight calculation model of weather forecast is:

[0142]

[0143] Wherein, DEQ is the second weight information, X and Y represent the forecast data information matrix and the analysis data information matrix respectively, T is the matrix transpose, and W is a preset N-row and N-column weight matrix; the expression of the weight matrix is:

[0144]

[0145] It should be noted that the preset weight matrix can effectively handle the uncertainty or importance of different observation data to reduce their impact on the fitting results and improve the overall accuracy of the fitting, thereby improving the robustness of the model.

[0146] S247, using an eighth weight calculation model for weather forecast, calculating and processing the target weight information, the first weight information, and the second weight information to obtain updated target weight information, and determining the updated target weight information as the target weight information;

[0147] Among them, the eighth weight calculation model of weather forecast is:

[0148] MB′=ε1·MB+ε2·DYQ+ε3·DEQ 0≤ε1,ε2,ε3≤1;

[0149] ε1+ε2+ε3=1;

[0150] Wherein, MB′ is the updated target weight information, MB is the target weight information, DYQ and DEQ are the first weight information and the second weight information, ε1, ε2 and ε3 are the first weight factor, the second weight factor and the third weight factor;

[0151] It should be noted that the first weight factor, the second weight factor and the third weight factor may be set by a user or obtained through historical data, and the embodiment of the present invention does not specifically limit this.

[0152] It should be noted that by calculating and processing the target weight information, the first weight information and the second weight information through the eighth weight calculation model of weather forecast, the advantages of different weight information can be effectively utilized, the deviation and variance of a single model can be reduced, and a more robust and accurate prediction result can be obtained. It can better adapt to unseen data, thereby improving the overall prediction accuracy.

[0153] It can be seen that the implementation of the weather forecast data processing method based on dynamic weights described in the embodiment of the present invention can obtain stable, comprehensive and integrated objective forecast result information through multi-mode forecast data information, which is beneficial to improving the meteorological forecast service capabilities.

[0154] In an optional embodiment, the step of processing the next time period forecast data information set according to the target weight information to obtain the target forecast information includes:

[0155] S31, preprocessing the next period forecast data information set to obtain a next period preprocessed forecast data information set; the next period preprocessed forecast data information set includes N next period preprocessed forecast data information;

[0156] It should be noted that the next time period is obtained by adding a time interval value to the end time point of the time window range in the time window range information. For example, if the time window range information is from 00:00 on June 30, 2024 to 00:00 on July 1, 2024, the next time period is 06:00 on July 1, 2024; the forecast data information set for the next time period is the GRAPE forecast data, ECMWF forecast data, GFS forecast data, and WRF forecast data obtained at 06:00 on July 1, 2024.

[0157] S32, using the fourth weight calculation model of weather forecast, calculating and processing the target weight information and the next period pre-processed forecast data information set to obtain target forecast information;

[0158] Wherein, the fourth weight calculation model of weather forecast is:

[0159]

[0160] Where, F′ is the target forecast information, F″ i1 is the i1th next period pre-processed forecast data information in the next period pre-processed forecast data information set, ω i1 is the i1th target forecast weight value in the target weight information, and N is the number of the next time period pre-processed forecast data information in the next time period pre-processed forecast data information set.

[0161] It can be seen that the implementation of the weather forecast data processing method based on dynamic weights described in the embodiment of the present invention can obtain stable, comprehensive and integrated objective forecast result information through multi-mode forecast data information, which is beneficial to improving the meteorological forecast service capabilities.

[0162] Embodiment 2

[0163] See also Figure 2 , Figure 2 is a schematic diagram of a weather forecast data processing device based on dynamic weights disclosed in an embodiment of the present invention. Figure 2 The described weather forecast data processing device based on dynamic weights is applied to a weather forecast data processing optimization system based on dynamic weights, such as a local server or cloud server for weather forecast data processing based on dynamic weights, and the embodiments of the present invention are not limited thereto. Figure 2 As shown, the weather forecast data processing device based on dynamic weights includes:

[0164] The acquisition module 201 is used to acquire a forecast data information set and analysis data information; the forecast data information set includes N forecast data information; N is a positive integer;

[0165] A first calculation module 202 is used to fuse the forecast data information set and the analysis data information to obtain target weight information;

[0166] The second calculation module 203 is used to process the next period forecast data information set according to the target weight information to obtain target forecast information; the next period forecast data information set includes N next period forecast data information.

[0167] It can be seen that the implementation of the dynamic weight-based weather forecast data processing device described in the embodiment of the present invention can obtain stable, comprehensive and integrated objective forecast result information through multi-mode forecast data information, which is beneficial to improving the meteorological forecast service capability.

[0168] Embodiment 3

[0169] See also Figure 3 , Figure 3is a schematic diagram of the structure of another weather forecast data processing device based on dynamic weights disclosed in an embodiment of the present invention. Figure 3 The described weather forecast data processing device based on dynamic weights is applied to a weather forecast data processing optimization system based on dynamic weights, such as a local server or cloud server for weather forecast data processing based on dynamic weights, and the embodiments of the present invention are not limited thereto. Figure 3 As shown, the weather forecast data processing device based on dynamic weights includes:

[0170] Processor 301;

[0171] A memory 302 coupled to the processor 301 and storing executable program codes;

[0172] The processor 301 calls the executable program code stored in the memory 302 to execute part or all of the steps of the weather forecast data processing method based on dynamic weights described in Example 1.

[0173] It can be seen that the implementation of the dynamic weight-based weather forecast data processing device described in the embodiment of the present invention can obtain stable, comprehensive and integrated objective forecast result information through multi-mode forecast data information, which is beneficial to improving the meteorological forecast service capability.

[0174] Embodiment 4

[0175] An embodiment of the present invention discloses a computer-readable storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute some or all steps of the weather forecast data processing method based on dynamic weights described in Example 1.

[0176] Embodiment 5

[0177] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute part or all of the steps in the weather forecast data processing method based on dynamic weights described in Example 1.

[0178] The system embodiments described above are only illustrative, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, i.e., they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative labor.

[0179] Through the specific description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution can be essentially or partly contributed to the prior art in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0180] Finally, it should be noted that the weather forecast data processing method and device based on dynamic weights disclosed in the embodiments of the present invention only disclose the preferred embodiments of the present invention, which are only used to illustrate the technical solution of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A weather forecast data processing method based on dynamic weights, characterized in that: The method comprises: S1, obtaining a forecast data information set and analysis data information; the forecast data information set includes N forecast data information; N is a positive integer; S2, fusing the forecast data information set and the analysis data information to obtain target weight information; S3, processing the next period forecast data information set according to the target weight information to obtain target forecast information; the next period forecast data information set includes N next period forecast data information.

2. The weather forecast data processing method based on dynamic weight according to claim 1, characterized in that: The step of fusing the forecast data information set and the analysis data information to obtain target weight information includes: S21, preprocessing the forecast data information set and the analysis data information to obtain a preprocessed forecast data information set and preprocessed analysis data information; the preprocessed forecast data information set includes the N preprocessed forecast data information; S22, processing the pre-processed forecast data information set and the pre-processed analysis data information to obtain a forecast data information set to be processed and analysis data information to be processed; the forecast data information set to be processed includes the N forecast data information to be processed; the forecast data information to be processed includes M forecast data information at time points to be processed; the analysis data information to be processed includes M analysis data information at time points to be processed; and M is a positive integer; S23, performing calculation processing on the forecast data information set to be processed and the analysis data information to be processed to obtain forecast threshold information to be processed and a forecast average threshold; S24, calculating and processing the forecast threshold information to be processed and the forecast average threshold to obtain target weight information; the target weight information includes the N target forecast weight values.

3. The weather forecast data processing method based on dynamic weights according to claim 2, characterized in that: The preprocessing of the forecast data information set and the analysis data information to obtain a preprocessed forecast data information set and preprocessed analysis data information includes: S211, performing data cleaning processing on the forecast data information set and the analysis data information to obtain a cleaned forecast data information set and cleaned analysis data information; S212, performing bilinear interpolation processing on the cleaned forecast data information set to obtain an interpolated forecast data information set; S213, performing boundary check and category consistency check processing on the interpolated forecast data information set and the cleaned analysis data information respectively to obtain a preprocessed forecast data information set and preprocessed analysis data information.

4. The weather forecast data processing method based on dynamic weights according to claim 2 is characterized in that: The processing of the preprocessed forecast data information set and the preprocessed analysis data information to obtain the forecast data information set to be processed and the analysis data information to be processed includes: S221, obtaining time window range information and time interval value; S222, performing a first judgment process on the preprocessing forecast data information set and the preprocessing analysis data information according to the time window range information to obtain a first preprocessing forecast data information set and a first preprocessing analysis data information; S223: Perform a second judgment process on the first preprocessed forecast data information set and the first preprocessed analysis data information according to the time window range information and the time interval value to obtain a forecast data information set to be processed and an analysis data information to be processed.

5. The weather forecast data processing method based on dynamic weights according to claim 2, characterized in that: The step of calculating and processing the forecast data information set to be processed and the analysis data information to be processed to obtain forecast threshold information to be processed and a forecast average threshold includes: S231, using a weather forecast first weight calculation model, calculating and processing the forecast data information set to be processed and the analysis data information to be processed to obtain forecast threshold information to be processed; Wherein, the weather forecast first weight calculation model is: Where, T is the forecast threshold information to be processed, T i is the i-th forecast threshold in the forecast threshold information to be processed, F i,j,k is the kth forecast data value in the jth forecast data information at the time point to be processed in the i-th forecast data information to be processed in the forecast data information set to be processed, O j,k is the kth analysis data value in the jth analysis data information to be processed at the time point to be processed in the analysis data information to be processed, L i,j is the number of forecast data values ​​in the forecast data information at the jth time point to be processed in the ith forecast data information to be processed in the forecast data information set to be processed, M is the number of forecast data information at the time point to be processed in the forecast data information to be processed, and N is the number of forecast data information to be processed in the forecast data information set to be processed; S232, using a second weather forecast weight calculation model, calculating and processing the forecast threshold information to be processed to obtain a forecast average threshold; Wherein, the second weight calculation model of weather forecast is: In the formula, is the forecast average threshold.

6. The weather forecast data processing method based on dynamic weights according to claim 2, characterized in that: The calculating and processing the forecast threshold information to be processed and the forecast average threshold to obtain target weight information includes: S241, determining whether any of the forecast thresholds in the forecast threshold information to be processed is greater than the forecast average threshold, and obtaining a threshold determination result; When the threshold judgment result is yes, determining the preset forecast weight value to be the target forecast weight value corresponding to the forecast threshold; When the threshold judgment result is no, executing S242; S242, using a third weather forecast weight calculation model, calculating and processing the forecast threshold, the forecast threshold information to be processed, and the forecast average threshold, to obtain a target forecast weight value corresponding to the forecast threshold; Wherein, the third weight calculation model of weather forecast is: Wherein, ω is the target forecast weight value corresponding to the forecast threshold, D is the forecast threshold, T is the forecast threshold information to be processed, and T b is the bth forecast threshold value in the forecast threshold information to be processed that is less than or equal to the average threshold value, and V is the number of the forecast threshold values ​​in the forecast threshold information to be processed that are less than or equal to the average threshold value.

7. The weather forecast data processing method based on dynamic weights according to claim 1, characterized in that: The step of processing the next time period forecast data information set according to the target weight information to obtain target forecast information includes: S31, preprocessing the next period forecast data information set to obtain a next period preprocessed forecast data information set; the next period preprocessed forecast data information set includes N next period preprocessed forecast data information; S32, using the fourth weight calculation model of weather forecast, calculating and processing the target weight information and the next period pre-processed forecast data information set to obtain target forecast information; Wherein, the fourth weight calculation model of weather forecast is: Where, F′ is the target forecast information, F″ i1 is the i1th next period pre-processed forecast data information in the next period pre-processed forecast data information set, ω i1 is the i1th target forecast weight value in the target weight information, and N is the number of the next time period pre-processed forecast data information in the next time period pre-processed forecast data information set.

8. A weather forecast data processing device based on dynamic weights, characterized in that: The device comprises: An acquisition module, used for acquiring a forecast data information set and analysis data information; the forecast data information set includes N forecast data information; N is a positive integer; A first calculation module is used to fuse the forecast data information set and the analysis data information to obtain target weight information; The second calculation module is used to process the next period forecast data information set according to the target weight information to obtain target forecast information; the next period forecast data information set includes N next period forecast data information.

9. A weather forecast data processing device based on dynamic weights, characterized in that: The device comprises: processor; a memory coupled to the processor and storing executable program code; The processor calls the executable program code stored in the memory to execute the weather forecast data processing method based on dynamic weights as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the weather forecast data processing method based on dynamic weights as described in any one of claims 1-7.

Images