WO2020103677A1

WO2020103677A1 - Method and device for processing meteorological element data of numerical weather prediction

Info

Publication number: WO2020103677A1
Application number: PCT/CN2019/115125
Authority: WO
Inventors: 丁宇宇; 董天仁; 王蔚青; 丁明月; 郭树锋; 张宇; 马文珍; 耿琴兰; 隆文喜; 苟晓侃; 杨凡; 李国栋
Original assignee: 国网青海省电力公司; 北京金风慧能技术有限公司; 国网青海省电力公司信息通信公司; 国家电网有限公司
Priority date: 2018-11-21
Filing date: 2019-11-01
Publication date: 2020-05-28
Also published as: CN109543295A; CN109543295B

Abstract

A method and a device for processing meteorological element data of a numerical weather prediction. Said method comprises: acquiring first meteorological element data predicted by a numerical weather prediction at a meteorological station (S102); acquiring second meteorological element data obtained by observation at the meteorological station (S104); selecting third meteorological element data corresponding to time interval coincidence region of the first meteorological element data and the second meteorological element data, and obtaining a first error correction coefficient according to the third meteorological element data (S106); according to the first error correction coefficient, determining, by means of an error conversion model, a second error correction coefficient at a prediction period of the current day (S108); and revising, according to the second error correction coefficient, the first meteorological element data, so as to obtain revised first meteorological element data (S110).

Description

Meteorological element data processing method and device for numerical weather forecast

This application requires the priority of the Chinese patent application with the application number 201811394322.3 submitted to the China Patent Office on November 21, 2018. The entire contents of this application are incorporated by reference in this application.

Technical field

The present application relates to the technical field of data processing, for example, to a method and device for meteorological element data processing of numerical weather forecast.

Background technique

Numerical weather forecast data such as wind speed and wind direction are used as inputs, and the forecasted meteorological elements are converted into wind farm output power prediction and photovoltaic output power prediction through prediction algorithms. Therefore, accurate prediction of numerical weather forecast data can provide important decision support for power dispatching, and is one of the important determinants of the prediction accuracy of new energy generation power.

Model post-processing methods for wind speed forecasting include model output statistics, Kalman filtering, Back Propagation (BP) neural network, and adaptive partial least squares. Among them, the most widely used method is Model Output Statistics (MOS).

Aiming at the problem that the numerical weather forecast in the related technology does not use the information of the nearby observation data as the forecast parameter, resulting in the low reliability of the meteorological forecast results, no effective solution has yet been proposed.

Summary of the invention

The embodiments of the present application provide a meteorological element data processing method and device for numerical weather forecasting, to at least solve that the numerical weather forecasting in related technologies does not use the information of nearby observation data as forecast parameters, resulting in low reliability of the obtained meteorological forecast results The problem.

In an embodiment, the present application provides a meteorological element data processing method for numerical weather forecasting, including: acquiring first meteorological element data predicted by a numerical weather forecast of a weather station, wherein the first meteorological element data is in Predicted data of the weather station acquired before the first time point of the current day from the second time point of the previous day; obtaining second meteorological element data observed at the meteorological station, wherein the second meteorological element The data is the data observed between the second time point of the previous day and the third time point of the day; the third corresponding to the area where the time interval coincides in the first meteorological element data and the second meteorological element data is selected Meteorological element data, and the first error correction coefficient is obtained according to the third meteorological element data; the second error correction coefficient of the current forecast period is determined by an error conversion model according to the first error correction coefficient, wherein the forecast period is A time period between the time when the first meteorological element data is obtained according to the numerical weather forecast and the time when the first meteorological element data is released; the first meteorological element data according to the second error correction coefficient Make revisions to get the revised first meteorological element data.

In an embodiment, the present application further provides a meteorological element data processing apparatus for numerical weather forecasting, including: a first acquiring unit configured to acquire first meteorological element data predicted by a numerical weather forecast of a weather station, wherein The first meteorological element data is the data predicted by the meteorological station acquired before the first time point of the current day from the second time point of the previous day; the second acquiring unit is set to acquire the observations obtained at the meteorological station The second meteorological element data of, wherein the second meteorological element data is data observed between the second time point of the previous day and the third time point of the day; the third acquisition unit is set to select the The first meteorological element data corresponds to the third meteorological element data corresponding to the region where the time interval coincides in the second meteorological element data, and a first error correction coefficient is obtained according to the third meteorological element data; the first determining unit is set to The first error correction coefficient determines the second error correction coefficient of the current forecast period through an error conversion model, wherein the forecast period is from the time when the first meteorological element data is obtained according to the numerical weather forecast to the first A time period between the time when the meteorological element data is released; a fourth acquisition unit configured to revise the first meteorological element data according to the second error correction coefficient to obtain the revised first meteorological element data.

In an embodiment, the present application further provides a storage medium, the storage medium includes a stored program, wherein the program executes the above method.

In an embodiment, the present application further provides a processor, the processor is configured to run a program, and the above method is executed when the program is running.

BRIEF DESCRIPTION

FIG. 1 is a flowchart of a method for processing meteorological element data of numerical weather forecast according to an embodiment of the present application;

2 is a schematic diagram of the relationship between meteorological element data and time according to an embodiment of the present application;

3 is a flowchart of another meteorological element data processing method of numerical weather forecast according to an embodiment of the present application;

FIG. 4 is a schematic diagram of a meteorological element data processing device for numerical weather forecast according to an embodiment of the present application.

detailed description

The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application. The described embodiments are only a part of the embodiments of the present application, but not all the embodiments.

The terms “first” and “second” in the description and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequential order. The data used in this way are interchangeable under appropriate circumstances so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and their various variations are intended to cover non-exclusive inclusions, for example, processes, methods, systems, products or devices that include a series of steps or units need not be limited to being clearly listed Those steps or units may include other steps or units that are not explicitly listed or inherent to these processes, methods, products or equipment.

In one embodiment, the MOS method generally takes multiple forecast variable values as the forecast factors, and uses the actual weather or meteorological elements at the time of forecast as the forecast amount, and selects different models of sampling from years of historical data, thereby seeking for Statistical relations and laws, and establish corresponding regression equations. The application of MOS technology in conventional weather forecast has been deeply researched and applied in practice. For example, the MOS forecast system has been established to provide a reference for daily short-term element forecasting; the use of MOS for fine wind forecasting has obtained results that are significantly higher than the original model ’s atmospheric mesoscale model (Mesoscale Model Version 5, MM5) model forecast level. Utilizing MOS technology to provide wind forecast; also applying MOS technology to the field of wind power forecasting, using model output multi-layer wind speed, air temperature and multiple turbulence parameters to establish multiple linear regression with wind farm wind speed to improve model forecast results.

The Perfect Prediction (PP) method is to first establish the statistical relationship between the large-scale circulation and local meteorological elements based on the observation data, and then replace the observed large-scale circulation information with the output of the numerical forecast for forecasting. This method assumes that the output value of the model is completely consistent with the measured value, that is, it believes that the numerical prediction is completely correct.

The basic idea of Kalman Filter (KF) algorithm is to use the state estimation value at the previous time and the observation value at the current time to obtain the optimal estimation of the state variable at the current time of the dynamic system, including the two steps of forecasting and analysis. In the forecast stage, the predicted value of the current mode state is generated according to the previous mode state. In the analysis stage, the observation data is introduced, and the mode state is re-analyzed using the minimum variance estimation method. As the model state forecast continues and new observation data is input one after another, this process continues to move forward, that is, the model integrates the state forecast forward over time, and when the observation data appears, according to the model prediction error covariance matrix The relative size between (known) and the covariance matrix of the observed error (known) derives the minimum variance estimate for the state.

Aggregate forecasting refers to a set of different forecasting results for the same effective forecasting time. Differences between multiple forecasts can provide information about the probability distribution of the forecasted quantity. Multiple forecasts in a set forecast can have different initial conditions, boundary conditions, parameter settings, and can even be generated with completely independent numerical weather forecast models . Aggregate forecasting is a method of starting with some initial values with little correlation and obtaining some forecasted values. This is a classic concept of collective forecasting. In addition to considering the initial value problem, the uncertainty and randomness of many physical processes in the numerical model (such as parameterization schemes, etc.) are also considered, and some predicted values are obtained. This is a new set of predictions.

The above-mentioned related technologies generally have the following defects:

1. The relevant technology only considers the error of historical observation data and historical forecast results, and does not consider the comparison error of nearby observation data to the forecast results of the day, so it is impossible to use the observation data that is near after each numerical forecast result is released;

2. The error characteristics of the numerical forecast results are different each time. Using the same model parameters to correct the errors can only reduce the common error between multiple forecasts, but cannot reduce the error of the single forecast itself.

Example one

According to an embodiment of the present application, a method embodiment of a meteorological element data processing method of numerical weather forecast is provided, and the steps shown in the flowchart of the accompanying drawings may be executed in a computer system such as a set of computer-executable instructions, and Although the logical sequence is shown in the flowchart, in some cases, the steps shown or described may be performed in an order different from here.

FIG. 1 is a flowchart of a meteorological element data processing method of numerical weather forecast according to an embodiment of the present application. As shown in FIG. 1, the meteorological element data processing method of numerical weather forecast includes: S102 to S110.

S102: Obtain first meteorological element data predicted by a numerical weather forecast of a meteorological station, where the first meteorological element data is data predicted by the meteorological station acquired before the first time point of the day from the second time point of the previous day .

For example, for meteorological station A, before 7 a.m. every day, obtain the weather forecast of the value reported by the meteorological station from 20:00 on the previous day, and the forecast result of a certain meteorological element data S is recorded as F _r , a In the embodiment, the data time resolution is 15 minutes.

Among them, numerical weather forecast: refers to the calculation of the equations of hydrodynamics and thermodynamics in the process of weather evolution based on the actual atmospheric conditions and under certain initial and boundary conditions, through large-scale computers for numerical calculations, to predict the atmosphere for a certain period in the future Methods of movement status and weather phenomena.

S104: Obtain second meteorological element data observed at a meteorological station, where the second meteorological element data is data observed between the second time point of the previous day and the third time point of the same day.

Similarly, for example, for meteorological station A, the corresponding meteorological element data obtained in real time from 20:00 to 6:00 of the previous day of the meteorological station is obtained before 7 a.m., and is recorded as _Or . In one embodiment, The data time resolution is also 15 minutes.

S106: Select the third meteorological element data corresponding to the region where the time interval coincides in the first meteorological element data and the second meteorological element data, and obtain the first error correction coefficient according to the third meteorological element data.

For example, it can be judged whether the area where the time interval of the meteorological element data predicted by the numerical weather forecast coincides with the observed meteorological element data is not less than 10 hours. If it is less than 10 hours, continue to obtain more observed meteorological element data until it coincides The area reaches 10 hours. In this embodiment, 10 hours is just an example, and the predetermined duration may be set according to actual needs.

S108: Determine the second error correction coefficient of the current forecast period through the error conversion model according to the first error correction coefficient, where the forecast period is from the time when the first meteorological element data is obtained according to the numerical weather forecast to the time when the first meteorological element data is released The time period between.

S110: Revise the first meteorological element data according to the second error correction coefficient to obtain the revised first meteorological element data.

Through this embodiment, the first meteorological element data predicted by the numerical weather forecast of the meteorological station can be obtained, where the first meteorological element data is the meteorological station acquired before the first time point of the day at the second time point of the previous day Forecast data; at the same time obtain the second meteorological element data observed at the meteorological station, where the second meteorological element data is the data observed between the second time point of the previous day and the third time point of the day; And select the third meteorological element data corresponding to the area where the time interval coincides in the first meteorological element data and the second meteorological element data, and obtain the first error correction coefficient according to the third meteorological element data; through the error conversion model according to the first error correction coefficient Determine the second error correction coefficient for the forecast period of the day, where the forecast period is the period between the time when the first meteorological element data is obtained according to the numerical weather forecast and the time when the first meteorological element data is released; and the second error correction The coefficient revises the first meteorological element data to obtain the revised first meteorological element data. Compared with the related technology, only the error of historical observation data and historical forecast results is considered, and the comparison error of nearby observation data to the forecast results of the day is not considered, so it is impossible to use the nearby observation data after each numerical forecast result is released; and The error characteristics of the numerical prediction results are different. Using the same model parameters to correct the errors can only reduce the common error between multiple predictions, but cannot reduce the shortcomings of the error of the single prediction itself. The meteorological element data processing method of numerical weather forecast can realize the purpose of using the observation data reported from the time of the forecast of the numerical weather forecast to the release period to correct the subsequent forecast results of the forecast, so as to fully utilize the nearby observation data ( That is, the information of the second meteorological element data) improves the technical effect of the accuracy of the revised forecast, and thus solves the problem that the numerical weather forecast in the related technology does not use the information of the nearby observation data as the forecast parameter, resulting in the low reliability of the weather forecast problem.

In S106, selecting the third meteorological element data corresponding to the overlapping area of the first meteorological element data and the second meteorological element data may include: determining whether the overlapping area of the first meteorological element data and the second meteorological element data meets the time interval A predetermined condition is obtained, and the judgment result is obtained, wherein the predetermined condition is that the total duration corresponding to the overlapping area of the first meteorological element data and the second meteorological element data is greater than the predetermined duration; when the judgment result is the first meteorological element data and the second meteorological element When the time interval overlapping area in the data satisfies the predetermined condition, the weather element data corresponding to the time interval overlapping area is used as the third weather element data.

In an embodiment, when the judgment result is that the area where the time interval overlaps in the first meteorological element data and the second meteorological element data does not satisfy the predetermined condition, the first meteorological element data and the second meteorological element data are selected to coincide in the time interval The third meteorological element data corresponding to the area includes: determining the time difference of the acquisition time of the second meteorological element data relative to the third time point; continuing to acquire the second meteorological element data based on the time difference until the time difference is zero.

In S106, obtaining the first error correction coefficient according to the third meteorological element data may include: one-to-one correspondence between the first meteorological element data and the second meteorological element data in the third meteorological element data corresponding to the time interval coincidence area according to time To obtain the third meteorological element data sequence; based on the third meteorological element data sequence, the first error correction coefficient is determined by a first formula, where the first formula is: e (t) = O _r (t)-(a · F _r (t) + b), t represents the t-th time point of the forecast period, e represents the error between the first meteorological element data and the second meteorological element data at the t-th time point, a and b represent the first error Correction coefficient, F _r (t) represents the first meteorological element data at the t-th time point, _Or (t) represents the second meteorological element data at the t-th time point; where, based on the third meteorological element data The sequence determines the first error correction coefficient through the first formula, including: determining the error between the first meteorological element data and the second meteorological element data corresponding to all time points in the time zone coincidence area through the first formula, In the case where the sum of all errors in the time interval coincidence area is the smallest, the first error correction coefficient is determined.

In an embodiment, the first meteorological element data and the second meteorological element data in the third meteorological element data satisfy the following relationship: _Or (t) = a · F _r (t) + b.

For example, the weather can intercept the first data element of numerical weather day time and the data F _r O _r overlapped portion of the second time meteorological elements day observation obtained, i.e., the day before the day of 6:00 to 20:00, and forecasts The obtained meteorological element data correspond to the meteorological element data obtained according to time one by one, recorded as _Fr (t), _Or (t), where t = 1 ... N, N is the number of observation data in the time period . According to the least square method, the following equation is established: _Or (t) = a · F _r (t) + b, the error is: e (t) = _Or (t)-(a · F _r (t) + b) When the overall error is the smallest, that is, ∑ _t e (t) ^{2 is the} smallest, the first error correction coefficients a and b are solved, and denoted as (a _r , b _r ).

As an optional embodiment, before determining the second error correction coefficient of the current forecast period through the error conversion model according to the first error correction coefficient, the method may further include: determining an error conversion model; wherein, determining the error conversion model includes: Obtain the historical forecast meteorological element data and historical observation meteorological element data in the historical time period; determine the first historical error revision coefficient in the predetermined time interval and the second historical error revision coefficient in the historical forecast time period ; Train the first historical error revision coefficient and the second historical error revision coefficient to obtain the error conversion model.

For example, you can obtain the historical forecast meteorological element data and historical observation meteorological element data in the historical time period, for example, you can obtain the daily forecast F _h and the observed O _h time series in the historical M days; follow the above calculation error correction coefficient method , Calculate the error correction coefficients (a _ri , b _ri ) for the first 10 hours of each day in the historical M days, and the error correction coefficients (a _fi , b _fi ) for the forecast period (eg 28-52 hours), where i represents the i day. Using the correction coefficients (a _ri , b _ri ) and (a _fi , b _fi ) of historical M days, an error conversion model can be trained. Establish the equation: a _f (i) = c · _ar (i), b _f (i) = d · _br (i), and the least square method can be used to find the coefficient (c, d) when the error is the smallest.

FIG. 2 is a schematic diagram of the relationship between meteorological element data and time according to an embodiment of the present application. As shown in FIG. 2, the real-time meteorological element data includes: real-time predicted meteorological element data (ie, first meteorological element data) obtained by numerical weather forecast, Observed real-time observation meteorological element data (ie, second meteorological element data). Among them, when the time period of real-time forecasting meteorological element data and real-time observation meteorological element data reaches 10 hours, the first error correction coefficients ( _ar , _br ) are determined according to the real-time forecasting meteorological element data and real-time observation meteorological element data; Use the historical forecast meteorological element data and historical observation meteorological element data in the historical time period to train to obtain an error conversion model, and obtain (a _ri , b _ri ), (a _fi , b _fi ) according to the error conversion model, and obtain the minimum error time Coefficient (c, d). Using the coefficient (c, d) when the error is the smallest to obtain the second error correction coefficient (a _f , b _f ) in combination with the first error correction coefficient (a _r , b _r ), F _a is the corrected forecast meteorological element data, F _f is the forecasted meteorological element data before the correction, and F _h and O _h are the time series corresponding to the predicted meteorological element data and the observed meteorological element data in the historical time period.

In S108, determining the second error correction coefficient of the current forecast period through the error conversion model according to the first error correction coefficient may include: determining the error revision coefficient under the minimum error condition through the error conversion model; according to the minimum error condition The error correction coefficient and the first error correction coefficient determine the second error correction coefficient.

For example, according to the error conversion model and the real-time error correction coefficients (a _r , b _r ) for the day, the error correction coefficients (a _f , b _f ) for the forecast period of the day can be obtained: a _f = c · a _r , b _f = D · _br , where c and d are conversion model coefficients calculated in S106. According to the error correction coefficient (that is, the second error correction coefficient) (a _f , b _f ) of the forecast period of the day, the revised forecast result of the day can be calculated: F _a (t) = a _f · F _f (t) + b _f , where F _a is the forecasted meteorological element data after correction, F _f is the forecasted meteorological element data before correction, and t is the t-th time point in the forecast period.

In the above S110, revising the first meteorological element data according to the second error correction coefficient to obtain the revised first meteorological element data includes: determining the revised first meteorological element data through the second formula, where the second formula Is: F _a (t) = a _f · F _f (t) + b _f , F _a is the revised first meteorological element data, F _f is the revised first meteorological element data, t is the forecast time period At the t-th time point, a _f and b _f represent the second error correction coefficient.

The meteorological element data processing method for numerical weather forecast provided in the embodiments of the present application will be described below with reference to the drawings.

FIG. 3 is a flowchart of another meteorological element data processing method for numerical weather forecast according to an embodiment of the present application. As shown in FIG. 3, meteorological element data (ie, second meteorological element data) and values obtained from real-time observation are obtained. After the meteorological element data (that is, the first meteorological element data) in the prediction result of the weather forecast, determine whether the coincident area in time between the meteorological element data obtained by real-time observation and the meteorological element data in the numerical weather forecast result is greater than or equal to 10 hours; When the judgment result is yes, calculate the real-time error correction coefficient (that is, the first error correction coefficient), and combine the error conversion module to correct the meteorological element data predicted by the numerical weather forecast; if the judgment result is no, continue Obtain meteorological element data obtained by real-time observation.

The method provided by the embodiment of the present application can use the numerical weather forecast results of the day and the observation results of the same period, and can be corrected based on the latest meteorological element data, which improves the correction effect of the forecast results of the numerical weather forecast, taking into account each time The relationship between the error of the predetermined time (for example, 10 hours) before the numerical weather forecast and the error of the forecast time period, using historical meteorological element data to establish an error conversion model, which improves the accuracy of the correction algorithm; there are no restrictions on the type of meteorological element data, which can be applied It is suitable for the correction of various meteorological factors such as wind speed, temperature, irradiance, etc.

Example 2

According to an embodiment of the present application, a numerical weather forecasting meteorological element data processing apparatus is also provided. The numerical weather forecasting meteorological element data processing apparatus of the present embodiment may be configured to execute the method provided by the present embodiment. The meteorological element data processing device for numerical weather forecast provided by the embodiment of the present application will be described below.

FIG. 4 is a schematic diagram of a meteorological element data processing apparatus for numerical weather forecast according to an embodiment of the present application. As shown in FIG. 4, the apparatus includes: a first acquiring unit 41, a second acquiring unit 43, and a third acquiring unit 45, The first determination unit 47 and the fourth acquisition unit 49. The device will be described below.

The first acquiring unit 41 is configured to acquire the first meteorological element data predicted by the numerical weather forecast of the meteorological station, wherein the first meteorological element data is the second meteorological station acquired before the first time point of the current day. Predicted data from time.

The second obtaining unit 43 is configured to obtain the second meteorological element data observed at the weather station, wherein the second meteorological element data is observed between the second time point of the previous day and the third time point of the same day data.

The third obtaining unit 45 is configured to select the third meteorological element data corresponding to the region where the time interval coincides between the first meteorological element data and the second meteorological element data, and obtain the first error correction coefficient according to the third meteorological element data.

The first determining unit 47 is configured to determine the second error correction coefficient of the current forecast period based on the first error correction coefficient through the error conversion model, where the forecast period is from the time when the first meteorological element data is obtained according to the numerical weather forecast to the first The time period between the time when the meteorological element data is released.

The fourth obtaining unit 49 is configured to revise the first meteorological element data according to the second error correction coefficient to obtain the revised first meteorological element data.

The first acquiring unit 41 in this embodiment may be configured to execute S102 in the embodiment of the present application, and the second acquiring unit 43 in this embodiment may be configured to execute S104 in the embodiment of the present application. The third acquiring unit 45 may be configured to execute S106 in the embodiment of the present application, the first determining unit 47 in the embodiment may be configured to execute S108 in the embodiment of the present application, and the fourth acquiring unit 49 in the embodiment may be configured To execute S110 in the embodiment of the present application. The above modules are the same as the examples and application scenarios implemented by the corresponding steps, but are not limited to the contents disclosed in the above embodiments.

In this embodiment, the first meteorological element data predicted by the numerical weather forecast of the meteorological station may be acquired by the first acquiring unit, where the first meteorological element data is the meteorological station acquired before the first time point of the day. The predicted data from the second time point of the day; at the same time, the second meteorological element data obtained at the meteorological station is obtained by the second acquisition unit, where the second meteorological element data is from the second time point of the previous day to the current day Observed data at the third time point; use the third acquisition unit to select the third meteorological element data corresponding to the area where the time interval coincides between the first meteorological element data and the second meteorological element data, and obtain the first error according to the third meteorological element data Correction coefficient; use the first determination unit to determine the second error correction coefficient of the current forecast period through the error conversion model according to the first error correction coefficient, where the forecast period is from the time when the first meteorological element data is obtained according to the numerical weather forecast to the first The time period between the time when the meteorological element data is released; and using the fourth acquisition unit, it is set to revise the first meteorological element data according to the second error correction coefficient to obtain the revised first meteorological element data.

Optionally, the third acquiring unit includes: a determination subunit configured to determine whether the overlapping region of the first meteorological element data and the second meteorological element data satisfies a predetermined condition to obtain a determination result, wherein the predetermined condition is the first meteorological condition The total duration corresponding to the overlapping area of the time interval between the element data and the second meteorological element data is greater than the predetermined duration; the first determining subunit is set to determine that the overlapping area of the time interval in the first meteorological element data and the second meteorological element data satisfies the judgment result In the case of a predetermined condition, the meteorological element data corresponding to the overlapping time zone is used as the third meteorological element data.

Optionally, the third acquiring unit further includes: a second determining subunit, configured to determine the second when the judgment result is that the overlapping region of the time interval in the first meteorological element data and the second meteorological element data does not satisfy the predetermined condition The acquisition time of the meteorological element data is relative to the time difference at the third time point; the first acquisition subunit is set to continue to acquire the second meteorological element data based on the time difference until the time difference is zero.

Optionally, the third acquiring unit includes: a second acquiring subunit configured to correspond to the first meteorological element data and the second meteorological element data in the third meteorological element data corresponding to the time interval coincidence area respectively according to time, The third meteorological element data sequence is obtained; the third determining subunit is set to determine the first error correction coefficient based on the third meteorological element data sequence through the first formula, where the first formula is: e (t) = O _r (t )-(a · F _r (t) + b), t represents the t-th time point of the forecast period, e represents the error of the first meteorological element data and the second meteorological element data at the t-th time point, a and b represents the first error correction coefficient, F _r (t) represents the first meteorological element data at the t-th time point, _Or (t) represents the second meteorological element data at the t-th time point; where, the third determiner The unit includes: a determination module configured to determine the error between the first meteorological element data and the second meteorological element data corresponding to all time points in the time interval coincidence area through the first formula, within the time interval coincidence area When the sum of all errors is the smallest, the first error correction coefficient is determined.

Optionally, the first meteorological element data and the second meteorological element data in the third meteorological element data satisfy the following relationship: _Or (t) = a · F _r (t) + b.

Optionally, the meteorological element data processing device of the numerical weather forecast further includes: a second determining unit, which is set to determine before determining the second error correction coefficient of the current forecast period through the error conversion model according to the first error correction coefficient Error conversion model; wherein, the second determination unit includes: a third acquisition subunit, which is set to acquire historical predicted meteorological element data and historical observation meteorological element data within the historical period; a fourth determination subunit, which is set to determine the historical period The first historical error revision coefficient within a predetermined time interval within each day, and the second historical error revision coefficient within the historical forecast period; the fourth acquisition subunit is set to revise the first historical error revision coefficient and the second historical error revision The coefficients are trained to obtain the error conversion model.

Optionally, the first determining unit includes: a fifth determining subunit, which is set to determine the error correction coefficient in the case of the smallest error through the error conversion model; and a sixth determining subunit, which is set according to the case in which the error is the smallest The error correction coefficient and the first error correction coefficient determine the second error correction coefficient.

Optionally, the fourth obtaining unit includes: a seventh determining subunit, which is set to determine the revised first meteorological element data through a second formula, where the second formula is: F _a (t) = a _f · F _f (t) + b _f , F _a is the revised first meteorological element data, F _f is the revised first meteorological element data, t is the t-th time point of the forecast period, a _f and b _{f are} the Two error correction coefficients.

The above-mentioned device includes a processor and a memory, and the above-mentioned first acquisition unit 41, second acquisition unit 43, third acquisition unit 45, first determination unit 47, fourth acquisition unit 49, etc. are all stored in the memory as program units, which are processed by The device executes the above-mentioned program units stored in the memory to realize the corresponding functions.

The above processor contains a core, and the core retrieves the corresponding program unit from the memory. One or more kernels may be set, and the first meteorological element data may be revised according to the second error correction coefficient by adjusting the kernel parameters to obtain revised first meteorological element data.

The above memory may include non-permanent memory, random access memory (RAM) and / or non-volatile memory in computer-readable media, such as read-only memory (ROM) or Flash memory (flash RAM), the memory includes at least one memory chip.

In an embodiment, a storage medium is also provided. The storage medium includes a stored program, where the program executes the above method.

In an embodiment, a processor is further provided, and the processor is configured to run a program, wherein the above method is executed when the program is running.

An embodiment of the present application also provides a device, which includes a processor, a memory, and a program stored on the memory and executable on the processor. When the processor executes the program, the following steps are implemented: acquiring the weather value of the weather station The first meteorological element data for forecasting and forecasting, where the first meteorological element data is the data predicted by the meteorological station obtained before the first time point of the day from the second time point of the previous day; The second meteorological element data, where the second meteorological element data is data observed between the second time point of the previous day and the third time point of the day; the first meteorological element data and the second meteorological element data are selected The third meteorological element data corresponding to the area where the time interval coincides, and the first error correction coefficient is obtained according to the third meteorological element data; the second error correction coefficient of the current forecast period is determined by the error conversion model according to the first error correction coefficient, where the forecast The time period is the time period between the time when the first meteorological element data is obtained according to the numerical weather forecast and the time when the first meteorological element data is released; the first meteorological element data is revised according to the second error correction coefficient, and the revised 1. Meteorological element data.

A computer program product is also provided in an embodiment of the present application. When executed on a data processing device, it is suitable for executing a program initialized with the following method steps: acquiring first meteorological element data predicted by a numerical weather forecast of a weather station, Among them, the first meteorological element data is the data predicted by the meteorological station obtained before the first time point of the day from the second time point of the previous day; the second meteorological element data obtained by observation at the meteorological station is obtained. The second meteorological element data is the data observed between the second time point of the previous day and the third time point of the day; the third meteorological element corresponding to the area where the time interval coincides between the first meteorological element data and the second meteorological element data is selected Element data, and the first error correction coefficient is obtained according to the third meteorological element data; the second error correction coefficient of the current forecast period is determined by the error conversion model according to the first error correction coefficient, where the forecast period is based on the numerical weather forecast A time period between the time of the first meteorological element data and the time when the first meteorological element data is released; the first meteorological element data is revised according to the second error correction coefficient to obtain the revised first meteorological element data.

The sequence numbers of the above embodiments of the present application are for description only, and do not represent the advantages and disadvantages of the embodiments.

In the above-mentioned embodiments of the present application, the descriptions of the multiple embodiments have respective emphasis. For a part that is not described in a certain embodiment, reference may be made to related descriptions of other embodiments.

In the several embodiments provided in this application, the disclosed technical content may be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the unit may be a logical function division. In actual implementation, there may be another division manner, for example, multiple units or components may be combined or Integration into another system, or some features can be ignored, or not implemented. The displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, units or modules, and may be in electrical or other forms.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, multiple functional units in multiple embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or software functional unit.

If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. All or part of the technical solution of the present application can be embodied in the form of a software product, which is stored in a storage medium and includes multiple instructions to make a computer device (which can be a personal computer, server, or network device) Etc.) to perform all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage media include: Universal Serial Bus flash disk (Universal Serial Bus flash disk, U disk), ROM, RAM, mobile hard disk, magnetic disk or optical disk and other media that can store program code.

Claims

A meteorological element data processing method for numerical weather forecast, including:

Obtain the first meteorological element data predicted by the numerical weather forecast of the meteorological station, wherein the first meteorological element data is predicted by the meteorological station acquired before the first time point of the current day from the second time point of the previous day The data;

Acquiring second meteorological element data observed at the weather station, wherein the second meteorological element data is observed between the second time point of the previous day and the third time point of the same day The data obtained;

Selecting the third meteorological element data corresponding to the region where the time interval coincides in the first meteorological element data and the second meteorological element data, and obtaining a first error correction coefficient according to the third meteorological element data;

The second error correction coefficient of the current forecast period is determined by an error conversion model according to the first error correction coefficient, where the forecast period is the time when the first meteorological element data is obtained according to the numerical weather forecast The time period between the time when the first meteorological element data is released;

Revise the first meteorological element data according to the second error correction coefficient to obtain revised first meteorological element data.
The method according to claim 1, wherein the selecting third meteorological element data corresponding to a region where time intervals coincide in the first meteorological element data and the second meteorological element data includes:

Judging whether the area where the time interval coincides in the first meteorological element data and the second meteorological element data satisfies a predetermined condition to obtain a judgment result, wherein the predetermined condition is the first meteorological element data and the second meteorological element The total duration corresponding to the coincident area of the time interval of the element data is greater than the predetermined duration;

When the result of the determination is that the time-region overlap region in the first meteorological element data and the second weather element data satisfies the predetermined condition, the meteorological element data corresponding to the time-region overlap region is used as the The third meteorological element data.
The method according to claim 2, after the obtaining the judgment result, further comprising: after the judgment result is that the overlapping area of time intervals in the first meteorological element data and the second meteorological element data does not satisfy the Under predetermined conditions, the selection of the third meteorological element data corresponding to the region where the time interval coincides in the first meteorological element data and the second meteorological element data includes:

Determine a time difference between the acquisition time of the second meteorological element data and the third time point;

Continue to acquire the second meteorological element data based on the time difference until the time difference is zero.
The method according to claim 1, wherein the obtaining the first error correction coefficient based on the third meteorological element data includes:

The first meteorological element data and the second meteorological element data in the third meteorological element data corresponding to the time interval coincidence area are respectively one-to-one corresponding to time to obtain a third meteorological element data sequence;

The first error correction coefficient is determined by a first formula based on the third meteorological element data sequence, where the first formula is: e (t) = O r (t)-(a · F r (t) + b), t represents the t-th time point of the forecast period, e represents the error between the first meteorological element data and the second meteorological element data at the t-th time point, a and b represent the first error correction coefficient, F r (t) represents the first meteorological element data at the t-th time point, Or (t) represents the t-th time Point second meteorological element data;

Wherein, the determining the first error correction coefficient by the first formula based on the third meteorological element data sequence includes:

The first formula is used to determine the error between the first meteorological element data and the second meteorological element data corresponding to all time points in the time interval coincidence area, and all the errors in the time interval coincidence area When the sum of errors is minimum, the first error correction coefficient is determined.
The method according to claim 4, wherein the first meteorological element data and the second meteorological element data in the third meteorological element data satisfy the following relationship: Or (t) = a · F r (t) + b .
The method according to claim 5, before determining the second error correction coefficient of the current forecast period through the error conversion model according to the first error correction coefficient, further comprising: determining the error conversion model;

Wherein, the determining the error conversion model includes:

Obtain historical forecast meteorological element data and historical observation meteorological element data within the historical time period;

Determining a first historical error revision coefficient within a predetermined time interval and a second historical error revision coefficient within a historical forecast period during the historical time period;

Training the first historical error revision coefficient and the second historical error revision coefficient to obtain the error conversion model.
The method according to claim 6, wherein the determining the second error correction coefficient of the current forecast period through the error conversion model according to the first error correction coefficient includes:

Determining the error correction coefficient under the condition of minimum error through the error conversion model;

The second error correction coefficient is determined according to the error correction coefficient with the smallest error and the first error correction coefficient.
The method according to claim 7, wherein the first meteorological element data is revised according to the second error correction coefficient to obtain revised first meteorological element data, including:

The revised first meteorological element data is determined by a second formula, where the second formula is: F a (t) = a f · F f (t) + b f , the F a is the revised The first meteorological element data of, F f represents the first meteorological element data before revision, t represents the t-th time point of the forecast period, the a f and the b f represent the second error Correction factor.
A meteorological element data processing device for numerical weather forecast, including:

The first acquiring unit is configured to acquire the first meteorological element data predicted by the numerical weather forecast of the meteorological station, wherein the first meteorological element data is the meteorological station acquired before the first time point of the day The predicted data from the second time on the day;

A second acquiring unit configured to acquire second meteorological element data observed at the weather station, wherein the second meteorological element data is from the second time point of the previous day to the current day Observed data between the third time point;

The third obtaining unit is configured to select the third meteorological element data corresponding to the region where the time interval coincides in the first meteorological element data and the second meteorological element data, and obtain a first error correction according to the third meteorological element data coefficient;

The first determining unit is configured to determine a second error correction coefficient of the current forecast period according to the first error correction coefficient through an error conversion model, wherein the forecast period is obtained by predicting the A time period between the time of one meteorological element data and the time when the first meteorological element data is released;

The fourth obtaining unit is configured to revise the first meteorological element data according to the second error correction coefficient to obtain revised first meteorological element data.
The apparatus according to claim 9, wherein the third acquiring unit comprises:

A judging subunit, configured to judge whether the overlapping area of the first meteorological element data and the second meteorological element data satisfies a predetermined condition to obtain a judgment result, wherein the predetermined condition is the first meteorological element data The total duration corresponding to the area where the time interval of the second meteorological element data coincides is greater than the predetermined duration;

The first determining subunit is configured to, when the judgment result is that the time-region overlapping region in the first meteorological element data and the second meteorological element data satisfies the predetermined condition, the time-region overlapping region The corresponding meteorological element data serves as the third meteorological element data.