CN106327005A - Air sounding data-based atmosphere three dimensional temperature field correction method - Google Patents

Abstract

The invention relates to an air sounding data-based atmosphere three dimensional temperature field correction method comprising the following steps: a zone where a three dimensional temperature field is positioned is divided according to a grid; each grid point is subjected to the following determining operation: if forecast air temperature and forecast wind speed of the grid point in a subsequent period of time are lower than a certain value, the following correction operation is conducted: temperature data is obtained at a certain elevation of the grid point at a number of different time points, temperature forecast data is correspondingly obtained via a conventional forecasting method, and temperature field correction coefficients are calculated; a correction matrix is obtained via correction coefficients of the grid point at different elevations; a future atmosphere three dimensional temperature field is forecast via the conventional forecasting method, a forecast result of each grid point is corrected via the correction matrix, and a corrected forecast result is formed. Via use of the method, high-altitude atmosphere three dimensional temperature field conditions can be corrected in a scientific and comprehensive manner, power grid icing forecast results can be corrected, and power grid icing forecast accuracy can be improved.

Description

A kind of air three-dimensional temperature field modification method based on sounding data

Technical field

The invention belongs to electrical engineering technical field, particularly relate to a kind of air three-dimensional temperature field based on sounding data and repair Correction method.

Background technology

Easily there is freezing rain and snow disaster in south China area, serious freezing rain and snow disaster once occurs, by electrical network Safe operation causes serious impact.For tackling Ice Disaster in Power Grids in advance, it is crucial for carrying out the prediction of electrical network icing.But, accurately Prediction electrical network whether the order of severity of ice damage and Ice Disaster in Power Grids can occur, it is necessary to accurately grasp upper atmosphere three-dimensional temperature field Situation, in conjunction with the temperature conditions of air three-dimensional temperature field, is calculated by numerical model, draws and whether icing can occur, and calculate Go out the electrical network ice covering thickness of follow-up 3～7 days.

One difficult problem of the prediction of air three-dimensional temperature field and detection always scientific circles.Although existing numerical weather calculates System can extrapolate the situation of the upper atmosphere three-dimensional temperature field of follow-up 3～7 days, but the meter of this air three-dimensional temperature field Calculate and there is bigger error, easily cause icing and predict the outcome bigger with practical situation difference.To hold air accurately three-dimensional The situation in temperature field, it is necessary to carry out the detection of upper atmosphere three-dimensional temperature field and revise work.

Summary of the invention

The technical problem to be solved in the present invention is: be not modified for traditional meteorological numerical forecast high and medium temperature forecast Problem, it is provided that a kind of air three-dimensional temperature field modification method based on sounding data, use the method can with science, comprehensively Revise upper atmosphere three-dimensional temperature field situation, correct electrical network icing forecast result, improve the accuracy rate of electrical network icing forecast.

The present invention, by detection differing heights, the temperature data of different location, forms the three-dimensional temperature field correction in this region Coefficient, revises the numerical forecast result of air three-dimensional temperature field, improves the forecast accuracy of air three-dimensional temperature field.

The technical scheme is that a kind of air three-dimensional temperature field modification method based on sounding data, including:

By three-dimensional temperature field region according to stress and strain model, following judgement carries out for each lattice point:

If the forecast temperature of this lattice point follow-up a period of time and forecast wind speed are respectively less than a certain value, then carry out following correction Process:

Some different time point at a certain elevation of this lattice point, it is thus achieved that temperature data, corresponding acquisition uses routine pre- The temperature forecast data that reporting method obtains, then calculate temperature field correction factor；

Correction matrix is obtained at the correction factor of different elevations by this lattice point；

Use the conventional following air three-dimensional temperature field of forecasting procedure forecast, for the forecast result of each lattice point, Jing Guoxiu The correction of positive matrices, forms the forecast result revised.

Further, it is judged that condition is within follow-up 24 hours, to forecast that temperature, whether below 5 DEG C, forecasts wind speed in follow-up 24 hours Whether less than 10 meter per seconds.

Further, it is thus achieved that each lattice point is in the temperature of 100,000 handkerchiefs, 92,500 handkerchiefs, 85,000 handkerchiefs, 70,000 handkerchiefs totally four elevations Degrees of data.

Further, the noise to the temperature data obtained uses Secondary Exponential Smoothing Method to process.

Further, arbitrary lattice point takes temperature T of two different time points at a certain elevation_i1、T_i2, use conventional pre- Temperature D that reporting method obtains_i1、D_i2, it is calculated as follows correction factor

$\begin{array}{cc}{a}_i=\frac{T_{i1}-T_{i2}}{D_{i1}-D_{i2}}& b_i=T_{i1}-D_{i1}\×\frac{T_{i1}-T_{i2}}{D_{i1}-D_{i2}}\end{array}.$

Further, arbitrary lattice point takes multiple time point at a certain elevation, uses method of least square to calculate correction factor.

Further, arbitrary lattice point uses the result of conventional forecasting procedure forecast to be d in the temperature field of a certain elevation_j, warp Crossing the forecast result revised is D_j, this lattice point is a at the correction factor of this elevation_j、b_j, wherein D_j=a_j×d_j+b_j。

The invention has the beneficial effects as follows:

1, use the technology of the present invention can air three-dimensional temperature field forecast data be modified, improve air three dimensional temperature The correctness of field prediction result, has filled up the technological gap of current shortage three-dimensional temperature field correction；

2, the air three-dimensional temperature field modification method that the present invention proposes and tradition uncorrected three-dimensional temperature field forecast data Comparing, temperature field data are less with real data error, and accuracy of the forecast is higher.

3, the air three-dimensional temperature field modification method principle that the present invention proposes is correct, clear process, strong operability, it is easy to real Existing, calculate process simple, can be that the forecast of electrical network icing provides data supporting, provide more accurate for the work of electrical network anti-ice Information of forecasting instruct.

4, the method that the present invention provides, improves the accuracy rate of electrical network icing forecast, can be power grid enterprises' science, efficient portion The time has been won in administration's electrical network anti-ice measure, improves the specific aim of the anti-icing work of electrical network, greatly reduces casualty loss, to electrical network Safe and stable operation played important function, the method promotion prospect is good.

Accompanying drawing explanation

Accompanying drawing 1 is region, temperature field stress and strain model.

Detailed description of the invention

In a detailed description of the invention, the detailed step of the air three-dimensional temperature field modification method of the present invention is as follows:

(1), air three-dimensional temperature field lattice point divides.

Three-dimensional temperature field region (can be adjusted according to required precision sizing grid according to the stress and strain model of 3km × 3km Whole), obtain the latitude and longitude coordinates that each grid lattice point is corresponding.Such as Fig. 1, take lattice point A11 and proceed by air three-dimensional temperature field and repair Just.

(2), judge whether air three-dimensional temperature field correction work entry condition is set up.

(2.1) conventional forecasting procedure is used to obtain lattice point A11 ground real time temperature, wind speed and following 24 hours forecast knots Really, it is judged that whether following condition is set up.

A, temperature condition judge.Judge that follow-up 24 hours forecast temperature of sensing point whether below 5 DEG C, is the most then made The judgement that temperature condition is set up；If it is not, then make the invalid judgement of temperature condition.

B, wind friction velocity judge.Judge that follow-up 24 hours forecast wind speed of sensing point whether less than 10 meter per seconds, are the most then made Go out the judgement that wind friction velocity is set up；If it is not, then make the invalid judgement of wind friction velocity.

(2.2) when A, B all set up, then air three-dimensional temperature field correction work is started；It is false when A, B have one Time, the most do not start air three-dimensional temperature field correction work.

(3), the temperature of the different elevation of detection.

The mode using conventional radar or balloon carries out sounding, respectively record lattice point A11 100,000 handkerchiefs, 92,500 The temperature data (can adjust according to the elevation number that required precision is chosen) of handkerchief, 85,000 handkerchiefs, 70,000 handkerchiefs totally four elevations.

(4), sounding data noise processes.

Collect the air three dimensional temperature field data of sounding passback, be directed in computer, carry out data validity inspection, for There are the data of substantially distortion, use Secondary Exponential Smoothing Method to process.

(5), air three-dimensional temperature field curve is drawn.

Use the drawing instrument of specialty, using the different height of elevation, temperature datas as X-axis, Y-axis data, draw this lattice The air three-dimensional temperature field change curve of point.

(6) correction factor of single lattice point, is asked for.

For lattice point A11, select two different time points, take twice temperature data T of 100,000 handkerchiefs₁₁、T₁₂, with often The air three-dimensional temperature field of rule Numerical Prediction System forecast is in temperature forecast data D of 100,000 handkerchiefs₁₁、D₁₂, calculate A11 lattice point Temperature field correction factor a₁、b₁。

$\begin{array}{cc}{a}_1=\frac{T_{11}-T_{12}}{D_{11}-D_{12}}& b_1=T_{11}-D_{11}*\frac{T_{11}-T_{12}}{D_{11}-D_{12}}\end{array}$

In order to improve the degree of accuracy of calculating, it is also possible at A11 lattice point, select more time point, detect multiple temperature, And use method of least square, calculate correction factor a₁₀₀₀、b₁₀₀₀。

(7) correction matrix of single lattice point is asked for.

Repeating step (6), calculating A11 lattice point is at 92,500 handkerchiefs, 85,000 handkerchiefs, the correction factor of 70,000 handkerchief height, and formation is repaiied Positive matrices forms correction matrix H_A11、H_B11。

$\begin{array}{cc}{H}_{A11}=\left(\begin{array}{c}{a}_{1000}\\ a_{925}\\ a_{850}\\ a_{700}\end{array}\right)& H_{B11}=\left(\begin{array}{c}{b}_{1000}\\ b_{925}\\ b_{850}\\ b_{700}\end{array}\right)\end{array}$

(8), the three-dimensional temperature field corrected Calculation of single lattice point.

Utilize convenient value Forecast Mode (WRF), the following air three-dimensional temperature field of forecast, for single lattice point A11,1000 The forecast result in hundred handkerchief temperature fields is d₁₀₀₀, then the forecast result through revision is D₁₀₀₀, wherein

D₁₀₀₀=a₁₀₀₀×d₁₀₀₀+b₁₀₀₀

Revised matrix is

$D_{A11}=\left(\begin{array}{c}{D}_{1000}\\ D_{925}\\ D_{850}\\ D_{700}\end{array}\right)$

(9), three-dimensional correction matrix is asked for.

At other lattice point, repeat step (2)～(6), calculate the correction matrix of all lattice points, form three-dimensional temperature field empty Between correction matrix H_A、H_B。

From now on, when calculating the related data of three-dimensional temperature field, it is possible to use above-mentioned three-dimensional correction matrix is repaiied Positive forecast data, improves accuracy of the forecast.

(10), three-dimensional correction result is asked for.

Calculate the temperature adjustmemt result of all lattice points, form the air three-dimensional temperature field revised.

$D=\left(\begin{array}{cccc}{D}_{A11}& D_{A12}& \mathrm{...}& D_{A1n}\\ D_{A21}& D_{A22}& \mathrm{...}& D_{A2n}\\ \mathrm{......}& \mathrm{...}& & \mathrm{...}\\ D_{An1}& D_{An2}& \mathrm{...}& D_{An2}\end{array}\right)$

Below by way of a specific embodiment, the present invention will be described in detail.

Embodiment 1:

As a example by choosing the typical easily area, little Sha river, Shaoyang City Longhui County, Ice Area that Hunan China saves, carry out air Three-dimensional temperature field correction.

(1), air three-dimensional temperature field lattice point divides.

Being divided according to the grid of 3km*3km in region, little Sha river, little Sha river land area is about 167.7 squares of public affairs In, 18 grids can be divided, obtain the latitude and longitude coordinates that each grid lattice point is corresponding, taking test site lattice point coordinate is east longitude 110 degree 46 points 33 seconds, north latitude 27 degree 31 points 39 seconds.

(2), judge whether air three-dimensional temperature field correction work entry condition is set up.

(2.1) real time temperature, wind velocity condition and the forecast result of testing site grid lattice point are collected, it is judged that following condition whether Set up.

A, temperature condition judge.Judge that follow-up 24 hours forecast temperature of sensing point, below 5 DEG C, is made temperature condition and set up Judgement.

B, wind friction velocity judge.Judge that follow-up 24 hours forecast wind speed of sensing point, less than 10 meter per seconds, make wind friction velocity Vertical judgement.

(2.2) A, B all set up, and start air three-dimensional temperature field correction work.

(3), the temperature of the different elevation of detection.

The mode using conventional radar or balloon carries out sounding, respectively testing site 100,000 handkerchiefs, 92,500 handkerchiefs, 850 The temperature data of hundred handkerchiefs, 70,000 handkerchiefs totally four elevations.

(4), sounding data noise processes.

Collect the air three dimensional temperature field data of sounding passback, be directed in computer, carry out data validity inspection, for Having the data of substantially distortion, the method using conventional secondary smooth processes.

(5), air three-dimensional temperature field curve is drawn.

Use the drawing instrument of specialty, using the different height of elevation, temperature datas as X-axis, Y-axis data, draw this lattice The air three-dimensional temperature field change curve of point.

(6) correction factor of single lattice point, is asked for.

For test site, select two different time points, take twice temperature data T of 100,000 handkerchiefs₁₁、T₁₂, with often The air three-dimensional temperature field of rule Numerical Prediction System forecast is in temperature forecast data D of 100,000 handkerchiefs₁₁、D₁₂, calculate experimental field Point temperature field correction factor a₁=1.05, b₁=0.058.

In order to improve the degree of accuracy of calculating, select more time point in test site, detect multiple temperature, and use Method of least square, finally calculates correction factor a₁₀₀₀Be 1.056, b₁₀₀₀It is 0.045.

(7) correction matrix of single lattice point is asked for.

Repeat step (6), calculate test site in 92,500 handkerchiefs, 85,000 handkerchiefs, the correction factor of 70,000 handkerchief height, formation Correction matrix forms correction matrix H_A11、H_B11。

$\begin{array}{cc}{H}_{A11}=\left(\begin{array}{c}1.056\\ 1.042\\ 0.995\\ 1.038\end{array}\right)& H_{B11}=\left(\begin{array}{c}0.045\\ 0.024\\ -0.016\\ 0.013\end{array}\right)\end{array}$

(8), the three-dimensional temperature field corrected Calculation of single lattice point.

Utilizing WRF Numerical Prediction Models, the following air three-dimensional temperature field of forecast, for single lattice point A11,100,000 handkerchief temperature The forecast result of degree field is d₁₀₀₀=-0.15 DEG C, then the forecast result through revision is D₁₀₀₀=-0.11 DEG C, actual observation result For-0.1 DEG C, wherein

D₁₀₀₀=a₁₀₀₀×d₁₀₀₀+b₁₀₀₀

Revised matrix is

Contrasting with actual observation, the accuracy that predicts the outcome of correction averagely improves 0.05 DEG C.

Claims (7)

1. an air three-dimensional temperature field modification method based on sounding data, it is characterised in that including:

By three-dimensional temperature field region according to stress and strain model, following judgement carries out for each lattice point:

If the forecast temperature of this lattice point follow-up a period of time and forecast wind speed are respectively less than a certain value, then carry out following correction Journey:

Some different time point at a certain elevation of this lattice point, it is thus achieved that temperature data, corresponding acquisition uses conventional forecast side The temperature forecast data that method obtains, then calculate temperature field correction factor；

Correction matrix is obtained at the correction factor of different elevations by this lattice point；

Use the conventional following air three-dimensional temperature field of forecasting procedure forecast, for the forecast result of each lattice point, through revising square The correction of battle array, forms the forecast result revised.

Air three-dimensional temperature field modification method based on sounding data the most according to claim 1, it is characterised in that: judge Condition is within follow-up 24 hours, to forecast that temperature, whether below 5 DEG C, forecasts that whether wind speed is less than 10 meter per seconds for follow-up 24 hours.

Air three-dimensional temperature field modification method based on sounding data the most according to claim 1, it is characterised in that: obtain Each lattice point is at the temperature data of 100,000 handkerchiefs, 92,500 handkerchiefs, 85,000 handkerchiefs, 70,000 handkerchiefs totally four elevations.

Air three-dimensional temperature field modification method based on sounding data the most according to claim 3, it is characterised in that: to obtaining The noise of the temperature data obtained uses Secondary Exponential Smoothing Method to process.

Air three-dimensional temperature field modification method based on sounding data the most according to claim 1, it is characterised in that: arbitrary Lattice point takes temperature T of two different time points at a certain elevation_i1、T_i2, use temperature D that conventional forecasting procedure obtains_i1、D_i2, It is calculated as follows correction factor

$\begin{array}{cc}{a}_i=\frac{T_{i1}-T_{i2}}{D_{i1}-D_{i2}}& b_i=T_{i1}-D_{i1}\×\frac{T_{i1}-T_{i2}}{D_{i1}-D_{i2}}\end{array}.$

Air three-dimensional temperature field modification method based on sounding data the most according to claim 1, it is characterised in that: arbitrary Lattice point takes multiple time point at a certain elevation, uses method of least square to calculate correction factor.

7. according to the air three-dimensional temperature field modification method based on sounding data described in claim 5 or 6, it is characterised in that: Arbitrary lattice point uses the result of conventional forecasting procedure forecast to be d in the temperature field of a certain elevation_j, through the forecast result revised it is D_j, this lattice point is a at the correction factor of this elevation_j、b_j, wherein D_j=a_j×d_j+b_j。