CN114091764A - Weather forecast element correction method, weather forecast element correction device, computer equipment and storage medium - Google Patents
Weather forecast element correction method, weather forecast element correction device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN114091764A CN114091764A CN202111406210.7A CN202111406210A CN114091764A CN 114091764 A CN114091764 A CN 114091764A CN 202111406210 A CN202111406210 A CN 202111406210A CN 114091764 A CN114091764 A CN 114091764A
- Authority
- CN
- China
- Prior art keywords
- meteorological
- weather forecast
- preset area
- grid
- mode
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The application relates to a meteorological forecast element correction method, a device, a computer device, a storage device and a computer program product, which can correct actual meteorological element data of a specific area and meteorological element data forecasted in a numerical mode, especially correct meteorological elements of areas which are easy to have extreme severe weather, such as coastal areas, and improve the accuracy of meteorological element prediction. The method comprises the following steps: acquiring a numerical weather forecast file which is output in a WRF mode and aims at a preset area; the numerical weather forecast file comprises the forecast values of various meteorological elements; respectively interpolating the predicted values of the various meteorological elements to longitude and latitude grid points of a preset area by using a nearest neighbor interpolation method to obtain an interpolation result; and inputting the interpolation result into a pre-constructed support vector machine model for calculation to obtain the revised value of each meteorological element.
Description
Technical Field
The present application relates to the field of weather prediction technologies, and in particular, to a method, an apparatus, a computer device, a storage device, and a computer program product for correcting weather forecast elements.
Background
The icing disaster of the power transmission line is one of the most serious meteorological disasters in the power system, and the safe and stable operation of a power grid is seriously threatened. In recent years, extreme weather frequently occurs, great pressure is brought to transmission line transportation facilities, and transmission line faults easily cause great loss to economy.
In order to reduce the loss of the transmission line caused by freezing disasters, the icing thickness of the transmission line is predicted accurately in advance, namely the icing thickness of the transmission line is predicted, wherein the icing thickness prediction of the transmission line refers to an information processing technology which is carried out by utilizing technologies such as a neural network and the like to carry out analysis and synthesis on a plurality of pieces of observation information obtained by a computer in time sequence on the temperature, the humidity, the wind speed, the wind direction, the air pressure, the sunshine, the precipitation quantity, the inclination angle and the wind deflection angle of an insulator and the weight of a wire acquired by a tension sensor under a certain rule so as to complete required decision and evaluation tasks. Wherein, the occurrence process of wire icing is inseparable from the change of meteorological elements. In recent years, a new generation of numerical Forecasting model, WRF (Weather Research and Forecasting model), has been widely used for Weather Forecasting.
However, the simulation effect of the WRF is affected by the terrain, the underlying surface, the resolution, the driving field and the physical process, and it is difficult to accurately predict meteorological elements based on a single WRF mode, and especially in areas with complex terrain, the simulation accuracy of the WRF mode is not accurate enough, and often fails to meet the actual requirements.
Disclosure of Invention
It is an object of the present invention to overcome the above-mentioned deficiencies of the prior art and to provide a method, an apparatus, a computer device, a computer readable storage medium and a computer program product for correcting weather forecast elements for accurately predicting weather elements at a coastal weather station.
In a first aspect, the present application provides a method for correcting weather forecast elements, the method comprising:
acquiring a numerical weather forecast file which is output in a WRF mode and aims at a preset area; the numerical weather forecast file comprises forecast numerical values of various meteorological elements;
respectively interpolating the predicted values of the multiple meteorological elements to longitude and latitude lattice points of the preset area by using a nearest neighbor interpolation method to obtain an interpolation result;
and inputting the interpolation result into a pre-constructed support vector machine model for calculation to obtain the revised value of each meteorological element.
In one embodiment, the method further comprises:
acquiring training sample data aiming at the preset area; the training sample data comprises predicted sample values of the meteorological elements and real sample values of the meteorological elements;
training a support vector machine model by using the training sample data, and adjusting model parameters in the support vector machine model by using a Bayesian parameter optimization algorithm to minimize a loss value between the real sample value and the predicted sample value, thereby obtaining the pre-constructed support vector machine model.
In one embodiment, the preset area comprises a plurality of meteorological sites; the interpolation method for the nearest neighbor is used for respectively interpolating the predicted values of the multiple meteorological elements to the longitude and latitude lattice points of the preset area to obtain an interpolation result, and the interpolation method comprises the following steps:
and finding out the grid point with the minimum distance by calculating the distance between the meteorological station and each longitude and latitude grid point, and assigning a prediction value corresponding to the grid point to the meteorological station to obtain the interpolation result.
In one embodiment, the obtaining of the numeric weather forecast file output in the WRF mode for the preset area includes:
acquiring forecast field data aiming at the preset area and earth surface static data aiming at the preset area;
and taking the forecast field data and the surface static data as the input of the WRF mode, defining the grid of the preset area as a 2-layer grid nesting mode, and performing simulation calculation on the meteorological model of the preset area by using a preset physical process parameterization scheme to generate a numerical weather forecast file of the preset area.
In one embodiment, the 2-layer mesh nesting mode comprises a first mesh nesting mode and a second mesh nesting mode; the grid number of the first grid nesting mode is 600 multiplied by 500, and the horizontal grid resolution of the first grid nesting mode is 9 km; the number of meshes of the second mesh nesting mode is 967 × 535, and the horizontal mesh resolution of the second mesh nesting mode is 3 km.
In one embodiment, the plurality of meteorological elements includes temperature, humidity, wind speed, precipitation, and barometric pressure.
In a second aspect, the application also provides a device for correcting the meteorological forecast elements. The device comprises:
the system comprises a numerical weather forecast file acquisition module, a data processing module and a data processing module, wherein the numerical weather forecast file acquisition module is used for acquiring a numerical weather forecast file which is output in a WRF mode and aims at a preset area; the numerical weather forecast file comprises the forecast numerical values of various meteorological elements;
the data interpolation module is used for respectively interpolating the predicted values of the various meteorological elements to longitude and latitude lattice points of the preset area by using a nearest neighbor interpolation method to obtain an interpolation result;
and the revision value calculation module is used for inputting the interpolation result into a pre-constructed support vector machine model for calculation to obtain the revision value of each meteorological element.
In a third aspect, the present application also provides a computer device. The computer device comprises a memory storing a computer program and a processor implementing the steps in the weather forecast element correction method embodiments when the processor executes the computer program.
In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps in the weather forecast element correction method embodiment.
In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program that implements the steps in the weather forecast element correction method embodiments when executed by a processor.
The weather forecast element correcting method, the weather forecast element correcting device, the computer equipment, the storage equipment and the computer program product are used for correcting the weather forecast element by acquiring a numerical weather forecast file which is output in a WRF mode and aims at a preset area; the numerical weather forecast file comprises the forecast values of various meteorological elements; respectively interpolating the predicted values of the various meteorological elements to longitude and latitude grid points of a preset area by using a nearest neighbor interpolation method to obtain an interpolation result; and inputting the interpolation result into a pre-constructed support vector machine model for calculation to obtain the revised value of each meteorological element. According to the method and the device, the meteorological elements of the specific area can be accurately predicted through a series of processing on the actual meteorological element data of the specific area and the meteorological element data of the numerical mode forecast, or the meteorological element results of the numerical mode forecast can be better corrected, especially, the meteorological elements can be corrected aiming at the areas which are easy to have extremely severe weather, such as coastal areas, scientific support is provided for the prediction of the icing condition of the power transmission lines of the areas, and the method and the device have important scientific significance and application value. Furthermore, the method and the device can be used for correcting the simulation result of the WRF mode in a targeted mode, so that meteorological data with higher precision can be obtained. By correcting meteorological elements of meteorological stations around the power transmission line tower, the meteorological elements at the power transmission line tower can be predicted more accurately, and therefore good early warning effect is achieved for ice coating prediction of the tower.
Drawings
FIG. 1 is a diagram illustrating an exemplary embodiment of a method for correcting weather forecast components;
FIG. 2 is a flow chart illustrating a method for correcting weather forecast elements according to an embodiment;
FIG. 3 is a diagram of the R of predicted values and actual values of WRF in a verification set, a test set and a WRF training set for predicting wind speed on towers of meteorological sites in one embodiment2(coefficient of determination), RMSE (root mean square error), MAE (mean absolute error), MAPE (mean percent absolute error) and R (coefficient of correlation) density maps;
FIG. 4 is a diagram of an actual value of the tower precipitation of the meteorological site, a predicted value of WRF, a correction value of the support vector machine, and a residual between an actual value of the wind speed and a predicted value of the support vector machine in one embodiment;
FIG. 5 is a block diagram showing a configuration of a weather forecast element correcting apparatus according to an embodiment;
FIG. 6 is a diagram illustrating an internal structure of a computer device according to an embodiment;
fig. 7 is an internal structural view of a computer device in another embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The method for correcting the weather forecast elements provided by the embodiment of the application can be applied to the application environment shown in fig. 1. Wherein the terminal 101 communicates with the server 102 via a network. The data storage system may store data that the server 102 needs to process. The data storage system may be integrated on the server 102, or may be located on the cloud or other network server. The terminal 101 may be, but not limited to, various sensors capable of detecting meteorological conditions or geological conditions, and may also be a personal computer, a notebook computer, a smart phone, a tablet computer, an internet of things device, and a portable wearable device, where the internet of things device may be an intelligent vehicle-mounted device, and the like. The portable wearable device can be a smart watch, a smart bracelet, a head-mounted device, and the like. The server 102 may be implemented as a stand-alone server or a server cluster comprising a plurality of servers.
In one embodiment, as shown in fig. 2, a method for correcting weather forecast elements is provided, which is illustrated by applying the method to the server 102 in fig. 1, and includes the following steps:
step S201, acquiring a numerical weather forecast file which is output in a WRF mode and aims at a preset area; the numerical weather forecast file comprises the forecast numerical values of various meteorological elements;
among them, WRF (Weather Research and Forecasting model) is a unified middle-scale Weather Forecasting model developed jointly by the american Environmental Forecasting Center (NCEP), the american National Atmospheric Research Center (NCAR), and multiple universities, Research institutes, and business departments, and is mainly characterized by advanced data assimilation technology, powerful nesting capability, and advanced physical processes, especially superior in convection and middle-scale precipitation handling capability. The WRF mode has wide application range, can be used for service numerical weather forecast, and can also be used in the field of atmospheric numerical simulation research, including research on data assimilation, research on physical process parameterization, regional climate simulation, air quality simulation, sea-air coupling, ideal experiment simulation and the like. The various meteorological factors include temperature, humidity, wind speed, precipitation, and barometric pressure.
Specifically, the WRF model is used for analyzing and predicting the current meteorological elements of a preset area (such as a coastal area) to obtain a numerical weather forecast file of the area in a future period. The numerical weather forecast file contains the predicted numerical values of various meteorological elements for each grid of the preset area. The WRF model is also a refined grid forecast model, similar to the transit network on the earth, and can decompose the preset area into a plurality of grids of 5 km × 5 km, even 1 km × 1 km, and the public lives in such individual grids, and the weather conditions in each grid are different. The WRF model carries out numerical simulation calculation on each grid, and compared with the traditional fixed-point prediction, the WRF model is more precise and more targeted in space. Regarding forecasting of the city A, the original forecast represents the weather condition of the whole city only by the temperature, precipitation and the like of a certain observation platform, but by developing WRF gridding forecast, the weather of the city A is not reflected by a fixed point any more, and the weather service and the weather forecast aiming at the city A can be finely reflected in different grids of the whole city.
Step S202, respectively interpolating the predicted values of the multiple meteorological elements to longitude and latitude grid points of the preset area by using a nearest neighbor interpolation method to obtain an interpolation result;
in particular, mathematically, the core idea of nearest neighbor interpolation is that each sample can be represented by its nearest neighbor. That is, by calculating the distance between the station and each grid point, the grid point with the smallest distance is found and the value of the grid point is assigned to the station. That is, the distance between the weather station and each longitude and latitude grid point is calculated, the grid point with the minimum distance is found out, and the prediction value corresponding to the grid point is assigned to the weather station, so that the grid point weather forecast element output in the WRF mode can be interpolated to the weather station with different longitude and latitude through the above description, and the interpolation result is obtained.
And step S203, inputting the interpolation result into a pre-constructed support vector machine model for calculation to obtain the revised value of each meteorological element.
Specifically, the previously constructed support vector machine model is a model obtained by training according to the difference between the past predicted value and the actual meteorological element value of the area, and the current predicted value of the area is input into the support vector machine model to be adjusted, so that the revised value of each meteorological element after revision can be obtained.
In the embodiment, the numerical weather forecast file output in the WRF mode and aiming at the preset area is acquired; the numerical weather forecast file comprises the forecast values of various meteorological elements; respectively interpolating the predicted values of the various meteorological elements to longitude and latitude grid points of a preset area by using a nearest neighbor interpolation method to obtain an interpolation result; and inputting the interpolation result into a pre-constructed support vector machine model for calculation to obtain the revised value of each meteorological element. The method can accurately predict the meteorological elements of the specific area through a series of processing on the actual meteorological element data of the specific area and the meteorological element data predicted in the numerical mode, or better correct the meteorological element result predicted in the numerical mode, especially correct the meteorological elements of the area which is easy to have extremely severe weather, such as coastal areas, and provide scientific support for the prediction of the icing condition of the power transmission line of the area, and has important scientific significance and application value.
In an embodiment, the method further includes: acquiring training sample data aiming at the preset area; the training sample data comprises predicted sample values of the meteorological elements and real sample values of the meteorological elements; training a support vector machine model by using the training sample data, and adjusting model parameters in the support vector machine model by using a Bayesian parameter optimization algorithm to minimize a loss value between the real sample value and the predicted sample value, thereby obtaining the pre-constructed support vector machine model.
Specifically, a specific application scenario is taken as an example, and selected from 8 at 10/2/2021 to 16 at 11/2/2021, 17.3 ° N-32.1 ° N in latitude, 89.6 ° E-119.4 ° E in longitude, 1 hour in time resolution, and 3km in lattice point resolution, the temperature, relative humidity, precipitation, air pressure, and 10m wind speed meteorological elements predicted in the WRF mode are interpolated to 45 meteorological stations with known longitudes and latitudes by using a nearest neighbor interpolation method, and 45 meteorological station monitoring data including temperature, relative humidity, precipitation, air pressure, and 10m wind speed in the same time are combined, and in addition, the known information of the stations is combined: and according to the dimension of time, combining WRF forecast data interpolated to the site and actual monitoring data of the site to form a training set and a testing set of a support vector machine algorithm, and removing abnormal values and null values. Here, we select the training set from 8/10/2/2021 to 15/10/29/2021 as the training set, and the test set from 5 days, i.e., 16/29/10/29/2021 to 16/11/2/11/2021 as the test set, where the test set does not participate in model training, the training set performs 5 cross experiments, and finds out the parameter setting of the support vector machine model with the highest precision in the training set by using the bayesian parameter optimization method, and the model parameters of the support vector machine include: c and gamma, the specific steps are as follows:
step A: utilizing a WRF mode, combining the data resolution of 0.25 degrees multiplied by 0.25 degrees per day, forecasting every 3 hours when the reporting time is 18 hours in the world, and taking the NCEP/GFS forecast field data frequently forecasted in 102 hours as the initial field and side boundary conditions of the WRF mode; obtaining land surface static data such as terrain, soil data, vegetation coverage and the like with the resolution of 15s (about 500m) based on an MODIS satellite; combining 2 layers of grid nesting layers, wherein the grid number is respectively 600 multiplied by 500 and 967 multiplied by 535, the horizontal grid resolution is respectively 9km and 3km, and the central point of the grid is arranged in WRF grids at 29-degree N and 96-degree E; in combination with a parameterized scheme named "CONUS": the micro-physics scheme is a Thompson scheme, the cloud accumulation parameterization scheme is a Tiedtke scheme, the long and short wave radiation schemes are RRTMG schemes, the boundary layer and the near-ground parameterization scheme are MYJ schemes, and the road process scheme adopts a Noah road process scheme to generate a WRFOUT numerical weather forecast file (comprising meteorological elements such as temperature, humidity, precipitation and the like). The specific grid arrangement is as in fig. 2.
And B: and (3) interpolating the forecasted ground meteorological elements of temperature, relative humidity, precipitation, air pressure and 10m wind speed meteorological elements in a WRF mode with the grid point resolution of 3km to 45 towers with known longitude and latitude by using a nearest neighbor interpolation method.
And C: actual monitoring data of 45 meteorological stations in the same time comprise temperature, relative humidity, precipitation, air pressure and 10m wind speed, and in addition, known information of the towers is combined: and (3) according to the altitude, combining WRF forecast data interpolated to the tower and actual monitoring data of the tower to form a training set and a test set of a support vector machine algorithm according to the dimension of time.
Step D: and finding out the parameter setting of the support vector machine algorithm with the highest precision of the training set by using a Bayesian parameter optimization method, wherein the specific parameters comprise C and gamma. Training on prediction of wind speed (temperature, air pressure, relative humidity and precipitation) on tower, verifying set, testing set and R of predicted value and actual value of WRF2The coefficients were 0.88, 0.83, 0.77 and-1.19, respectively, as shown in FIG. 3. And the actual value of the tower precipitation of the meteorological station, the predicted value of the WRF, the correction value of the support vector machine, and the residual error between the actual value of the wind speed and the predicted value of the support vector machine are shown in FIG. 4.
The model for the above-described support vector machine model is explained as follows: for a general regression problem, a given training sample D { (x)1,y1),(x2,y2),...,(xn,yn)},yiE.g., R, we wish to learn a f (x) so that it is in contact withy are as close as possible and w, b are the parameters to be determined. In this model, the loss is zero only if f (x) is identical to y, and support vector regression assumes that we can tolerate a maximum deviation of ε between f (x) and y, and the loss is calculated if and only if f (x) differs from y by an absolute value greater than ε, which is equivalent to constructing a 2 ε -wide interval band centered on f (x), and if the training samples fall within this interval band, it is considered to be predicted correctly. (the slack on either side of the spacer tape can be different)
In summary, the loss function metric of the SVM regression model is:
defining an SVM regression model objective function as follows:
s.t.|yi-w·φ(xi)-b|≤ε(i＝1,2,…,m) (3)
the regression model may also add a relaxation variable to each sample, but since the absolute values are used here at s.t., there are actually two inequalities, i.e., both sides need the relaxation variable, which is defined as
The loss function of the SVM regression model after adding the relaxation variable is:
introducing an inequality and a lagrange multiplier yields a lagrange function:
lagrange multiplier:
lagrange function:
traversing derivation, making the partial derivative zero, and obtaining:
substituting the above formula, eliminating w, b,
and taking the negative sign of the target function to obtain:
the above process satisfies the KKT condition, then:
finally, the solution of the SVR is obtained as:
the Bayesian parameter optimization method comprises the following steps:
the Bayesian optimization algorithm has two core contents, namely a probability agent model (PSM) and an acquisition function (AC), wherein the probability agent model models uncertainty of parameters according to a probability framework and comprises a prior probability model and an observation model, and the probability agent model observes prior probability to obtain posterior probability distribution comprising more prior; the acquisition function is composed of posterior probability distribution obtained by observed data, and the next parameter evaluation point with the most potential is obtained by maximizing the acquisition function.
The objective of the bayesian optimization algorithm is to minimize the objective function value, which can be expressed by equation 15:
f(x)＝Loss(Tv,x)+ε (16)
loss (Tc, x) may be represented by the following pseudo code:
def Loss(Tc,x):
a transfer learning model (containing a hyperparameter x);
training a training set;
the verification set Tv verifies the output Loss;
return logloss
f (x) -an objective function; tv-validation set; loss () -Loss function; x-hyperparameters; x-hyperparametric domain space.
Because the model hyperparameter x is in an optimization process and is always in an optimization state before the algorithm is terminated, a real objective function f (x) is unknown, when the Bayesian optimization algorithm is terminated, the hyperparameter reaches the optimum, and the objective function is known and is the obtained optimization model.
The probability agent model is divided into a parametric model and a nonparametric model according to whether the model has a fixed parameter set or not, the commonly used parametric model comprises a beta-Bernoulli model, a linear model and a generalized linear model, the nonparametric model comprises a Gaussian process, a random forest, a deep learning network and a Tree Park Estimators (TPE) method, and the TPE method supports hyper-parameters with a specified domain space, so that the TPE is adopted to construct the probability model.
The TPE method differs from other methods in that TPE does not construct a predictive probability model for the objective function f (x), but a probability model is generated for all domain variables using the density of equation 13.
y*＝min{f(xt),1≤t<n, which is the optimal value after H is observed; h { (x)i,f(xi) I is more than or equal to 1 and less than or equal to t) as a historical observation set of f (x); l (x) is such that the corresponding loss f (x)i) Less than y*The density formed; g (x) is such that the corresponding loss f (x)i) Y is not less than*The density formed.
The common acquisition functions are classified into three types, and are based on the lifting strategies, namely PI and EI; strategy based on information, including Thompson sampling, entropy searching sampling and entropy predicting sampling; a confidence bound policy and a combination policy. An Extended Improvement (EI) method based on a lifting strategy is adopted as an acquisition function and is optimized; EI is a desired function for mapping x to a real space R, and has the advantages of having few parameters and balancing exploration (exploration) and development (exploration) to some extent.
The final goal of the algorithm is to minimize the objective function f (x), so we want the evaluation point x returned by the collection function each time to reduce f (x) as much as possible, i.e. the distribution of x is located in the region of l (x) as much as possible, so we construct EI to explore and exploreThe development is balanced, i.e. the evaluation point is located in the l (x) area as much as possible, and a new evaluation point x can be obtained to enrich the l (x) area. Based on the above description, let β ═ p (y)<y*) I.e. by
Then:
substituting equations 19, 20 into 18 yields:
to minimize the objective function f (x), i.e. we want x to be as located as far as possible in the region l (x), so that β is larger, so by maximizing
Thereby obtaining an evaluation point x (i).
In summary, in the embodiment, the trained support vector machine model is obtained by training the support vector machine model through the training sample set of the preset region, so that the weather forecast elements of the coastal weather station can be corrected and predicted more accurately, the accuracy of weather forecast of the coastal weather station in the future and the probability of occurrence of a disaster in the future can be increased to a certain extent, and the method has important scientific significance and application value.
In an embodiment, the step S201 includes: acquiring forecast field data aiming at the preset area and surface static data aiming at the preset area; and taking the forecast field data and the surface static data as the input of the WRF mode, defining the grid of the preset area as a 2-layer grid nesting mode, and performing simulation calculation on the meteorological model of the preset area by using a preset physical process parameterization scheme to generate a numerical weather forecast file of the preset area. Wherein the 2-layer mesh nesting mode comprises a first mesh nesting mode and a second mesh nesting mode; the number of grids in the first grid nesting mode is 600 x 500, and the horizontal grid resolution of the first grid nesting mode is 9 km; the number of meshes of the second mesh nesting mode is 967 × 535, and the horizontal mesh resolution of the second mesh nesting mode is 3 km.
Specifically, the information of the NCEP/GFS (National Centers for Environmental Prediction, National center for Environmental Prediction of America) Forecast field with the daily data resolution of 0.25 degrees multiplied by 0.25 degrees and the Forecast time of every 3 hours of the world 18 is obtained, and the Forecast field data of 102 hours total is taken as the initial field and side boundary conditions of the WRF mode; acquiring land surface static data such as terrain, soil data, vegetation coverage and the like with the Resolution of 15s (about 500m) provided by a MODIS (Moderate-Resolution Imaging spectrometer) -based satellite (namely a satellite provided with an MODIS sensor); (ii) a This is the data that the parameters contain in the WRF mode. Combining 2 layers of grid nesting layers, wherein the grid number is respectively 600 multiplied by 500 and 967 multiplied by 535, the horizontal grid resolution is respectively 9km and 3km, and the central point of the grid is arranged in WRF grids at 29-degree N and 96-degree E; input can set parameters in the WRF mode. In combination with a parameterized scheme named "CONUS": the micro-physics scheme is a Thompson scheme, the cloud accumulation parameterization scheme is a Tiedtke scheme, the long and short wave radiation schemes are RRTMG schemes, the boundary layer and the near-ground parameterization scheme are MYJ schemes, and the pavement process scheme adopts a Noah pavement process scheme to generate a WRFOUT numerical weather forecast file (comprising meteorological elements such as temperature, humidity, precipitation and the like); input can set parameters in the WRF mode. Interpolating the weather forecast data of the WRFOUT grid point to the longitude and latitude positions of the weather station by using a nearest neighbor method; acquiring five meteorological elements of temperature, humidity, wind speed, precipitation and air pressure of a coastal meteorological site and the altitude of the site; matching the forecast data and the actual observation data in time to serve as a training and testing data set for supporting a vector machine model; and finding out the optimal parameters for training the support vector machine model by using a Bayesian parameter optimization method, and predicting the temperature, humidity, wind speed, precipitation and air pressure of the meteorological site on a test set.
According to the embodiment, the actual meteorological element data of the coastal meteorological site and the meteorological element data of the numerical mode forecast are obtained, and through a series of processing, the meteorological elements of the coastal meteorological site can be accurately forecasted or the meteorological element results of the numerical mode forecast can be better corrected, scientific support is provided for the transmission line icing occurrence forecast, and the method has important scientific significance and application value.
Optionally, the actual meteorological element data of the coastal meteorological site include, but are not limited to, temperature, relative humidity, precipitation, 10m wind speed, and surface air pressure. WRF forecast data includes, but is not limited to, 2m temperature, relative humidity, precipitation, 10m wind speed, and surface air pressure.
It should be understood that, although the steps in the flowcharts related to the embodiments as described above are sequentially displayed as indicated by arrows, the steps are not necessarily performed sequentially as indicated by the arrows. The steps are not limited to being performed in the exact order illustrated and, unless explicitly stated herein, may be performed in other orders. Moreover, at least a part of the steps in the flowcharts related to the embodiments described above may include multiple steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the execution order of the steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a part of the steps or stages in other steps.
Based on the same inventive concept, the embodiment of the application also provides a weather forecast element correcting device for realizing the weather forecast element correcting method. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the method, so the specific limitations in the embodiment of the weather forecast element correction device or devices provided below can be referred to the limitations of the weather forecast element correction method in the above, and details are not repeated herein.
In one embodiment, as shown in fig. 5, there is provided a weather forecast element correcting device 500 including: a numerical weather forecast file obtaining module 501, a data interpolation module 502 and a revised value calculating module 503, wherein:
a numerical weather forecast file acquiring module 501, configured to acquire a numerical weather forecast file for a preset area, where the numerical weather forecast file is output in the WRF mode; the numerical weather forecast file comprises forecast numerical values of various meteorological elements;
the data interpolation module 502 is configured to interpolate the predicted values of the multiple meteorological elements to longitude and latitude grid points of the preset area by using a nearest neighbor interpolation method, so as to obtain an interpolation result;
and a revised value calculation module 503, configured to input the interpolation result into a pre-constructed support vector machine model for calculation, so as to obtain a revised value of each meteorological element.
In an embodiment, the apparatus further includes a model training unit, configured to obtain training sample data for the preset region; the training sample data comprises predicted sample values of the meteorological elements and real sample values of the meteorological elements; training a support vector machine model by using the training sample data, and adjusting model parameters in the support vector machine model by using a Bayesian parameter optimization algorithm to minimize a loss value between the real sample value and the predicted sample value, thereby obtaining the pre-constructed support vector machine model.
In an embodiment, the preset area comprises a plurality of meteorological sites; the data interpolation module 502 is further configured to: and finding out the grid point with the minimum distance by calculating the distance between the meteorological station and each longitude and latitude grid point, and assigning a prediction value corresponding to the grid point to the meteorological station to obtain the interpolation result.
In an embodiment, the above-mentioned numerical weather forecast file obtaining module 501 is further configured to: acquiring forecast field data aiming at the preset area and earth surface static data aiming at the preset area; and taking the forecast field data and the surface static data as the input of the WRF mode, defining the grid of the preset area as a 2-layer grid nesting mode, and performing simulation calculation on the meteorological model of the preset area by using a preset physical process parameterization scheme to generate a numerical weather forecast file of the preset area.
In an embodiment, the 2-layer mesh nesting mode comprises a first mesh nesting mode and a second mesh nesting mode; the grid number of the first grid nesting mode is 600 multiplied by 500, and the horizontal grid resolution of the first grid nesting mode is 9 km; the number of meshes of the second mesh nesting mode is 967 × 535, and the horizontal mesh resolution of the second mesh nesting mode is 3 km.
In one embodiment, the plurality of meteorological elements includes temperature, humidity, wind speed, precipitation, and barometric pressure.
All or part of the modules in the weather forecast element correcting device can be realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.
In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 6. The computer device includes a processor, a memory, and a network interface connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer equipment is used for storing data such as meteorological element predicted values and real values. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of correcting a weather forecast component.
In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as shown in fig. 7. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program to run on the non-volatile storage medium. The communication interface of the computer device is used for carrying out wired or wireless communication with an external terminal, and the wireless communication can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. The computer program is executed by a processor to implement a method of correcting a weather forecast component. The display screen of the computer equipment can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer equipment can be a touch layer covered on the display screen, a key, a track ball or a touch pad arranged on the shell of the computer equipment, an external keyboard, a touch pad or a mouse and the like.
It will be appreciated by those skilled in the art that the configurations shown in fig. 6-7 are only block diagrams of some of the configurations relevant to the present disclosure, and do not constitute a limitation on the computing devices to which the present disclosure may be applied, and that a particular computing device may include more or less components than shown in the figures, or may combine certain components, or have a different arrangement of components.
In one embodiment, a computer device is provided, which comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the steps of the weather forecast element correction method embodiment when executing the computer program.
In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which when executed by a processor implements the steps in the above-described weather forecast element correction method embodiments.
In one embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of the above-described weather forecast element correction method embodiment.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, database, or other medium used in the embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high-density embedded nonvolatile Memory, resistive Random Access Memory (ReRAM), Magnetic Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (PCM), graphene Memory, and the like. Volatile Memory can include Random Access Memory (RAM), external cache Memory, and the like. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), among others. The databases referred to in various embodiments provided herein may include at least one of relational and non-relational databases. The non-relational database may include, but is not limited to, a block chain based distributed database, and the like. The processors referred to in the various embodiments provided herein may be, without limitation, general purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, or the like.
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.
Claims (10)
1. A method for correcting a weather forecast element, comprising:
acquiring a numerical weather forecast file which is output in a WRF mode and aims at a preset area; the numerical weather forecast file comprises the forecast numerical values of various meteorological elements;
respectively interpolating the predicted values of the multiple meteorological elements to longitude and latitude lattice points of the preset area by using a nearest neighbor interpolation method to obtain an interpolation result;
and inputting the interpolation result into a pre-constructed support vector machine model for calculation to obtain the revised value of each meteorological element.
2. The method of claim 1, further comprising:
acquiring training sample data aiming at the preset area; the training sample data comprises predicted sample values of the meteorological elements and real sample values of the meteorological elements;
training a support vector machine model by using the training sample data, and adjusting model parameters in the support vector machine model by using a Bayesian parameter optimization algorithm to minimize a loss value between the real sample value and the predicted sample value, thereby obtaining the pre-constructed support vector machine model.
3. The method of claim 1, wherein the predetermined area includes a plurality of weather stations; the interpolation method for the nearest neighbor is used for respectively interpolating the predicted values of the multiple meteorological elements to the longitude and latitude lattice points of the preset area to obtain an interpolation result, and the interpolation method comprises the following steps:
and finding out the grid point with the minimum distance by calculating the distance between the meteorological station and each longitude and latitude grid point, and assigning a prediction value corresponding to the grid point to the meteorological station to obtain the interpolation result.
4. The method of claim 1, wherein the obtaining of the numeric weather forecast file for the preset area output in the WRF mode comprises:
acquiring forecast field data aiming at the preset area and earth surface static data aiming at the preset area;
and taking the forecast field data and the surface static data as the input of the WRF mode, defining the grid of the preset area as a 2-layer grid nesting mode, and performing simulation calculation on the meteorological model of the preset area by using a preset physical process parameterization scheme to generate a numerical weather forecast file of the preset area.
5. The method of claim 4, wherein the 2-layer mesh nesting pattern comprises a first mesh nesting pattern and a second mesh nesting pattern; the grid number of the first grid nesting mode is 600 multiplied by 500, and the horizontal grid resolution of the first grid nesting mode is 9 km; the number of meshes of the second mesh nesting mode is 967 × 535, and the horizontal mesh resolution of the second mesh nesting mode is 3 km.
6. The method of any one of claims 1 to 5, wherein the plurality of meteorological elements includes temperature, humidity, wind speed, precipitation, and barometric pressure.
7. A weather forecast element correction device, comprising:
the system comprises a numerical weather forecast file acquisition module, a data processing module and a data processing module, wherein the numerical weather forecast file acquisition module is used for acquiring a numerical weather forecast file which is output in a WRF mode and aims at a preset area; the numerical weather forecast file comprises the forecast numerical values of various meteorological elements;
the data interpolation module is used for respectively interpolating the predicted values of the various meteorological elements to longitude and latitude lattice points of the preset area by using a nearest neighbor interpolation method to obtain an interpolation result;
and the revision value calculation module is used for inputting the interpolation result into a pre-constructed support vector machine model for calculation to obtain the revision value of each meteorological element.
8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method of any of claims 1 to 6.
9. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program realizes the steps of the method of any one of claims 1 to 6 when executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111406210.7A CN114091764A (en) | 2021-11-24 | 2021-11-24 | Weather forecast element correction method, weather forecast element correction device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111406210.7A CN114091764A (en) | 2021-11-24 | 2021-11-24 | Weather forecast element correction method, weather forecast element correction device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114091764A true CN114091764A (en) | 2022-02-25 |
Family
ID=80304195
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111406210.7A Pending CN114091764A (en) | 2021-11-24 | 2021-11-24 | Weather forecast element correction method, weather forecast element correction device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114091764A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115391746A (en) * | 2022-10-28 | 2022-11-25 | 航天宏图信息技术股份有限公司 | Interpolation method, device, electronic device and medium for meteorological element data |
| CN115796394A (en) * | 2023-02-01 | 2023-03-14 | 天翼云科技有限公司 | Numerical weather forecast correction method and device, electronic equipment and storage medium |
| CN116070676A (en) * | 2023-03-28 | 2023-05-05 | 南京气象科技创新研究院 | Expressway road surface temperature forecasting method based on attention mechanism and self-encoder |
| CN116167509A (en) * | 2022-07-20 | 2023-05-26 | 中科三清科技有限公司 | Air temperature prediction method, air temperature prediction device, storage medium and electronic equipment |
| CN116186493A (en) * | 2023-04-24 | 2023-05-30 | 航天宏图信息技术股份有限公司 | Meteorological forecasting method, device, equipment and medium based on numerical mode correction |
| CN116578825A (en) * | 2022-12-28 | 2023-08-11 | 上海勘测设计研究院有限公司 | Meteorological prediction error correction method, device, medium and electronic equipment |
| CN116975731A (en) * | 2023-08-08 | 2023-10-31 | 山东大学 | Cross-domain cutter damage monitoring method and system based on transfer learning |
-
2021
- 2021-11-24 CN CN202111406210.7A patent/CN114091764A/en active Pending
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116167509A (en) * | 2022-07-20 | 2023-05-26 | 中科三清科技有限公司 | Air temperature prediction method, air temperature prediction device, storage medium and electronic equipment |
| CN116167509B (en) * | 2022-07-20 | 2023-09-29 | 中科三清科技有限公司 | Air temperature prediction method, air temperature prediction device, storage medium and electronic equipment |
| CN115391746A (en) * | 2022-10-28 | 2022-11-25 | 航天宏图信息技术股份有限公司 | Interpolation method, device, electronic device and medium for meteorological element data |
| CN115391746B (en) * | 2022-10-28 | 2023-01-31 | 航天宏图信息技术股份有限公司 | Interpolation method, interpolation device, electronic device and medium for meteorological element data |
| CN116578825A (en) * | 2022-12-28 | 2023-08-11 | 上海勘测设计研究院有限公司 | Meteorological prediction error correction method, device, medium and electronic equipment |
| CN115796394A (en) * | 2023-02-01 | 2023-03-14 | 天翼云科技有限公司 | Numerical weather forecast correction method and device, electronic equipment and storage medium |
| CN116070676A (en) * | 2023-03-28 | 2023-05-05 | 南京气象科技创新研究院 | Expressway road surface temperature forecasting method based on attention mechanism and self-encoder |
| CN116070676B (en) * | 2023-03-28 | 2023-06-30 | 南京气象科技创新研究院 | Expressway road surface temperature forecasting method based on attention mechanism and self-encoder |
| CN116186493A (en) * | 2023-04-24 | 2023-05-30 | 航天宏图信息技术股份有限公司 | Meteorological forecasting method, device, equipment and medium based on numerical mode correction |
| CN116975731A (en) * | 2023-08-08 | 2023-10-31 | 山东大学 | Cross-domain cutter damage monitoring method and system based on transfer learning |
| CN116975731B (en) * | 2023-08-08 | 2024-02-20 | 山东大学 | Cross-domain cutter damage monitoring method and system based on transfer learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114091764A (en) | Weather forecast element correction method, weather forecast element correction device, computer equipment and storage medium | |
| Suparta et al. | Modeling of tropospheric delays using ANFIS | |
| Woldemeskel et al. | Merging gauge and satellite rainfall with specification of associated uncertainty across Australia | |
| Zeng et al. | Daily global solar radiation in China estimated from high‐density meteorological observations: a random forest model framework | |
| Cao et al. | Spatial sequential modeling and predication of global land use and land cover changes by integrating a global change assessment model and cellular automata | |
| US20190139059A1 (en) | Demand forecasting device, demand forecasting method and non-transitory computer readable medium | |
| CN104239706A (en) | Preparation method for ground-based observation air temperature space-time data set | |
| CN110618473A (en) | Instant meteorological information guarantee method | |
| CN110472782A (en) | A kind of data determination method, device, equipment and storage medium | |
| Li et al. | Possible impact of climate change on apple yield in Northwest China | |
| Osah et al. | Deep learning model for predicting daily IGS zenith tropospheric delays in West Africa using TensorFlow and Keras | |
| EP3794386B1 (en) | Determination of location-specific weather information for agronomic decision support | |
| Palmate et al. | Assessment of uncertainties in modelling land use change with an integrated cellular automata–Markov chain model | |
| Kheradmanda et al. | Clearness index predicting using an integrated artificial neural network (ANN) approach | |
| CN113222019A (en) | Meteorological forecast data processing method, device and equipment for power transmission line tower | |
| Gharaylou et al. | Prediction of lightning activity using WRF-ELEC model: Impact of initial and boundary conditions | |
| Fan et al. | A comparative study of four merging approaches for regional precipitation estimation | |
| Anuchaivong et al. | Statistical downscaling for rainfall forecasts using modified constructed analog method in Thailand | |
| Zhang et al. | Merging multisatellite precipitation products using stacking method and the censored-shifted gamma ensemble model output statistics in china's Beimiaoji basin | |
| CN112285808B (en) | Method for reducing scale of APHRODITE precipitation data | |
| Padhee et al. | Spatiotemporal reconstruction of MODIS land surface temperature with the help of GLDAS product using kernel-based nonparametric data assimilation | |
| Gao et al. | A novel dynamic interpolation method based on both temporal and spatial correlations | |
| CN116609859A (en) | Weather disaster high-resolution regional mode forecasting system and method | |
| Li et al. | Multivariate interpolation and information entropy for optimizing raingauge network in the Mekong River Basin | |
| CN114463616A (en) | Multi-source satellite precipitation fusion method based on Stacking and EMOS-CSG |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |