阅读说明：本技术 一种塔测风的多层风向风速概率分布计算方法 (Multilayer wind direction and wind speed probability distribution calculation method for tower wind measurement ) 是由 施萧 冯箫 贵志成 张晓杰 甘思旧 聂于棚 于 2021-07-06 设计创作，主要内容包括：本发明涉及一种塔测风的多层风向风速概率分布计算方法,属于环境风场分析领域。本发明为得到多层风向风速分布规律,以测风塔10米、50米、90米三层测风数据为例,利用混合von mises分布拟合风向、gamma分布拟合风速,再根据风向风速相关性得到逐层风向风速联合概率。随后通过matplolib三维绘图,建立测风塔多层风向风速概率分布模型。该方法可以科学认识多层风分布规律,适用于风能资源评估、工程结构设计,也可为浅层风预报提供基础。(The invention relates to a tower anemometry multilayer wind direction and wind speed probability distribution calculation method, and belongs to the field of environmental wind field analysis. In order to obtain a multilayer wind direction and wind speed distribution rule, taking three layers of wind measurement data of a wind measurement tower, namely 10 meters, 50 meters and 90 meters as examples, fitting wind direction and gamma distribution by utilizing mixed von mises distribution to fit wind speed, and obtaining a layer-by-layer wind direction and wind speed joint probability according to the correlation of wind direction and wind speed. And then, establishing a multilayer wind direction and wind speed probability distribution model of the anemometer tower through matplolib three-dimensional drawing. The method can scientifically know the multilayer wind distribution rule, is suitable for wind energy resource assessment and engineering structure design, and can also provide a foundation for shallow wind forecasting.)

1. A tower anemometry multilayer wind direction and wind speed probability distribution calculation method is characterized by comprising the following steps:

s1, layer-by-layer wind direction and wind speed probability distribution:

s11, processing data, and abandoning a calm wind field;

s12, fitting wind speed probability distribution;

s13, fitting wind direction probability distribution;

s2, wind direction and wind speed joint probability distribution:

s21, calculating wind direction and wind speed cumulative probability;

s22, calculating a wind direction and a wind speed correlation coefficient;

s23, calculating the wind direction and wind speed joint probability;

s3, a multilayer wind direction and wind speed probability distribution model:

s31, processing wind direction and wind speed joint probability data of each layer;

s32, multi-layer wind direction and wind speed probability distribution;

and S33, distributing the probability distance between wind directions and wind speeds in multiple layers.

2. The method for calculating the probability distribution of the multilayer wind directions and the wind speeds of the tower anemometry according to claim 1, wherein the step S11 specifically comprises: aggregating the minute wind field data hour by hour to obtain maximum wind speed hour by hour of 10 meters, 50 meters and 90 meters and corresponding wind direction; and the wind less than 0.5m/s is defined as a static wind field, and the data of the static wind field is discarded.

3. The method for calculating the probability distribution of the multilayer wind directions and the wind speeds of the tower anemometry according to claim 2, wherein the step S12 specifically comprises: the probability density function of the gamma distribution is:

wherein v is the wind speed, η is the shape parameter, and γ is the scale parameter;

and calculating shape parameters and scale parameters of gamma distribution according to the maximum likelihood estimation method for the wind speed data of each layer to obtain a corresponding wind speed distribution probability density function.

4. The method for calculating the probability distribution of the multilayer wind directions and the wind speeds of the tower anemometry according to claim 2, wherein the step S13 specifically comprises:

the probability density function of the mixed von mises distribution is:

where θ is the wind direction, i is the number of von mises distribution mixing groups, and w_iDistributing the weight occupied by each component; mu.s_iLocation parameter, k, for von mises distribution_iIs a scale parameter; i is₀(k_i) Is a modified zero order first class Bessel function;

calculating a probability density function of the mixed von mises distribution in an iterative mode;

according to the common wind direction, the number of the mixing groups is set to be 8; namely dividing the angle of 0-360 degrees into 8 directions at equal intervals; calculating the proportion of each azimuth wind direction data sample in the total wind direction sample, and taking the proportion as a probability accumulation value of the wind direction data in each azimuth, namely weight w;

calculating parameters of each group of von mises distribution by adopting a maximum likelihood estimation method; calculating the average sine value corresponding to each group of wind direction dataAnd cosine valueFurther obtaining the position parameters mu of each group of distribution;

at the mean sine valueAnd cosine valueOn the basis, a scale parameter k is obtained according to an empirical function;

if the weight w of each group is close to or equal to 0, the group is removed, and the number of the rest groups is calculated again by w, mu and k.

5. The method for calculating the probability distribution of the multilayer wind directions and the wind speeds for tower wind measurement according to any one of claims 1 to 4, wherein the step S21 specifically comprises: on the basis of the probability density function, the cumulative probability densities F (theta) and F (upsilon) of the wind direction and the wind speed are calculated through integration.

6. The method for calculating the probability distribution of the multilayer wind directions and the wind speeds of the tower anemometry according to claim 5, wherein the step S22 specifically comprises:

calculating sine values and cosine values of all wind direction data one by one to obtain sine values and cosine values of all wind direction data; then calculating a wind direction and wind speed correlation coefficient;

the calculation formula of the wind direction and wind speed correlation coefficient is as follows:

wherein r is_vcIs the correlation coefficient, r, of each wind speed with the corresponding wind direction cosine value_vsFor the correlation coefficient, r, of each wind speed with the sine value of the corresponding wind direction_csThe correlation coefficient between the wind direction cosine value and the sine value is obtained; the correlation coefficient of the wind direction and the wind speed is r.

7. The method for calculating the probability distribution of the multilayer wind directions and the wind speeds of the tower anemometry according to claim 6, wherein the step S23 specifically comprises:

the wind direction and wind speed joint probability distribution function is as follows:

f(ν,θ)＝f(ν)f(θ)[1+r(1-2F(υ)(1-2F(θ))]

v is the wind speed, and theta is the wind direction; f (nu) is a wind speed probability density function, and f (theta) is a wind direction probability density function; r is a wind direction and wind speed correlation coefficient; f (upsilon) is a wind speed cumulative density function, and F (theta) is a wind direction cumulative density function; and substituting the corresponding elements and functions into a formula to obtain a data set of wind direction and wind speed joint probability distribution.

8. The method for calculating the probability distribution of the multilayer wind directions and the wind speeds of the tower anemometry according to claim 7, wherein the step S31 specifically includes: the maximum value of the wind direction and wind speed joint probability f (v, theta) of each layer of 10 meters, 50 meters and 90 meters is obtained, and the maximum value is used as the display interval delta h of each layer in the Z-axis direction in the multilayer wind direction and wind speed probability distribution modeling.

9. The method for calculating the probability distribution of the multilayer wind directions and the wind speeds of the tower wind measurement according to claim 8, wherein the step S32 specifically comprises: taking wind direction and wind speed joint probabilities f (v, theta) of layers of 10 meters, 50 meters and 90 meters as input, carrying out three-dimensional drawing by utilizing matplotlib of python, and carrying out three-dimensional surface drawing by utilizing a plot _ surface function; the method comprises the steps of drawing 10 meters on a first layer, drawing 50 meters on a second layer at intervals of delta h, and drawing a third layer at intervals of delta h, namely 90 meters to obtain the multi-layer wind direction and wind speed joint probability.

10. The method for calculating the probability distribution of the multilayer wind directions and the wind speeds of the tower wind measurement according to claim 8, wherein the step S33 specifically comprises: averaging the wind direction and wind speed combined probabilities f (v, theta) of three layers of 10 meters, 50 meters and 90 meters to obtain a combined probability average valueThen respectively subtracting the probability average value by using the joint probabilities of 10 meters, 50 meters and 90 meters to obtain the joint probability interval of each layer; three-dimensional drawing is carried out by utilizing matplotlib of python, three-dimensional surface drawing is carried out by utilizing a plot _ surface function, 10 meters of drawing is carried out on a first layer, 50 meters of drawing is carried out at intervals delta h to serve as a second layer, and a third layer, namely 90 meters of drawing is carried out at intervals delta h to obtain the equal distribution of wind direction and wind speed joint probabilities of a multilayer wind field.

Technical Field

The invention belongs to the field of environmental wind field analysis, and particularly relates to a tower anemometry multilayer wind direction and wind speed probability distribution calculation method.

Background

In engineering application and meteorological guarantee, evaluation and analysis based on long-time wind field data are more critical, and the wind direction and wind speed analysis of the current multilayer wind field is dispersive, or single-layer analysis, or only single element of the wind speed or the wind direction is analyzed, and the wind direction and the wind speed probability distribution of the multilayer wind field is rarely involved. The problem can be solved by the method for calculating the probability distribution of the wind direction and the wind speed of the multilayer anemometry.

Disclosure of Invention

Technical problem to be solved

The invention aims to solve the technical problem of how to provide a tower anemometry multilayer wind direction and wind speed probability distribution calculation method so as to solve the problems of wind direction and wind speed separation and single analysis level in multilayer environment wind field analysis.

(II) technical scheme

In order to solve the technical problem, the invention provides a tower anemometry multilayer wind direction and wind speed probability distribution calculation method, which comprises the following steps:

s1, layer-by-layer wind direction and wind speed probability distribution:

s11, processing data, and abandoning a calm wind field;

s12, fitting wind speed probability distribution;

s13, fitting wind direction probability distribution;

s2, wind direction and wind speed joint probability distribution:

s21, calculating wind direction and wind speed cumulative probability;

s22, calculating a wind direction and a wind speed correlation coefficient;

s23, calculating the wind direction and wind speed joint probability;

s3, a multilayer wind direction and wind speed probability distribution model:

s31, processing wind direction and wind speed joint probability data of each layer;

s32, multi-layer wind direction and wind speed probability distribution;

and S33, distributing the probability distance between wind directions and wind speeds in multiple layers.

Further, the step S11 specifically includes: aggregating the minute wind field data hour by hour to obtain maximum wind speed hour by hour of 10 meters, 50 meters and 90 meters and corresponding wind direction; and the wind less than 0.5m/s is defined as a static wind field, and the data of the static wind field is discarded.

Further, the step S12 specifically includes: the probability density function of the gamma distribution is:

wherein v is the wind speed, η is the shape parameter, and γ is the scale parameter;

and calculating shape parameters and scale parameters of gamma distribution according to the maximum likelihood estimation method for the wind speed data of each layer to obtain a corresponding wind speed distribution probability density function.

Further, the step S13 specifically includes:

the probability density function of the mixed von mises distribution is:

where θ is the wind direction, i is the number of von mises distribution mixing groups, and w_iDistributing the weight occupied by each component; mu.s_iLocation parameter, k, for von mises distribution_iIs a scale parameter; i is₀(k_i) Is a modified zero order first class Bessel function;

calculating a probability density function of the mixed von mises distribution in an iterative mode;

according to the common wind direction, the number of the mixing groups is set to be 8; namely dividing the angle of 0-360 degrees into 8 directions at equal intervals; calculating the proportion of each azimuth wind direction data sample in the total wind direction sample, and taking the proportion as a probability accumulation value of the wind direction data in each azimuth, namely weight w;

calculating parameters of each group of von mises distribution by adopting a maximum likelihood estimation method; calculating the average sine value corresponding to each group of wind direction dataAnd cosine valueFurther obtaining the position parameters mu of each group of distribution;

at the mean sine valueAnd cosine valueOn the basis, a scale parameter k is obtained according to an empirical function;

if the weight w of each group is close to or equal to 0, the group is removed, and the number of the rest groups is calculated again by w, mu and k.

Further, the step S21 specifically includes: on the basis of the probability density function, the cumulative probability densities F (theta) and F (upsilon) of the wind direction and the wind speed are calculated through integration.

Further, the step S22 specifically includes:

calculating sine values and cosine values of all wind direction data one by one to obtain sine values and cosine values of all wind direction data; then calculating a wind direction and wind speed correlation coefficient;

the calculation formula of the wind direction and wind speed correlation coefficient is as follows:

wherein r is_vcIs the correlation coefficient, r, of each wind speed with the corresponding wind direction cosine value_vsFor the correlation coefficient, r, of each wind speed with the sine value of the corresponding wind direction_csThe correlation coefficient between the wind direction cosine value and the sine value is obtained; the correlation coefficient of wind direction and wind speed is r。

Further, the step S23 specifically includes:

the wind direction and wind speed joint probability distribution function is as follows:

f(ν,θ)＝f(ν)f(θ)[1+r(1-2F(υ)(1-2F(θ))]

v is the wind speed, and theta is the wind direction; f (nu) is a wind speed probability density function, and f (theta) is a wind direction probability density function; r is a wind direction and wind speed correlation coefficient; f (upsilon) is a wind speed cumulative density function, and F (theta) is a wind direction cumulative density function; and substituting the corresponding elements and functions into a formula to obtain a data set of wind direction and wind speed joint probability distribution.

Further, the step S31 specifically includes: the maximum value of the wind direction and wind speed joint probability f (v, theta) of each layer of 10 meters, 50 meters and 90 meters is obtained, and the maximum value is used as the display interval delta h of each layer in the Z-axis direction in the multilayer wind direction and wind speed probability distribution modeling.

Further, the step S32 specifically includes: taking wind direction and wind speed joint probabilities f (v, theta) of layers of 10 meters, 50 meters and 90 meters as input, carrying out three-dimensional drawing by utilizing matplotlib of python, and carrying out three-dimensional surface drawing by utilizing a plot _ surface function; the wind direction and the wind speed are drawn at 10 meters on the first layer, then 50 meters are drawn at an interval delta h as the second layer, and then the third layer, namely 90 meters is drawn at an interval delta h, so that the multilayer wind direction and wind speed joint probability is obtained.

Further, the step S33 specifically includes: averaging the wind direction and wind speed combined probabilities f (v, theta) of three layers of 10 meters, 50 meters and 90 meters to obtain a combined probability average valueThen respectively subtracting the probability average value by using the joint probabilities of 10 meters, 50 meters and 90 meters to obtain the joint probability interval of each layer; three-dimensional drawing is carried out by utilizing matplotlib of python, three-dimensional surface drawing is carried out by utilizing a plot _ surface function, 10 meters of drawing is carried out on a first layer, 50 meters of drawing is carried out at intervals delta h to serve as a second layer, and a third layer, namely 90 meters of drawing is carried out at intervals delta h to obtain the equal distribution of wind direction and wind speed joint probabilities of a multilayer wind field.

(III) advantageous effects

The invention provides a tower anemometry multilayer wind direction and wind speed probability distribution calculation method, which can obtain the probability distribution of multilayer wind direction and wind speed. Taking heights of 10 meters, 50 meters and 90 meters as examples, the multi-layer wind direction and wind speed probability distribution is realized by fitting the independent probabilities of the wind direction and the wind speed, calculating the wind direction and wind speed combined probability distribution, establishing a multi-layer wind direction and wind speed distribution model and the like. The method can evaluate and analyze the wind field more scientifically, and is suitable for wind energy resource evaluation, engineering structure design and shallow wind field forecasting.

Drawings

FIG. 1 shows the wind direction and speed probabilities of the tower for measuring 10 m, 50 m and 90 m wind;

FIG. 2 shows that the tower of the present invention measures wind with 10 m, 50 m and 90 m wind direction and wind speed probability distance.

Detailed Description

In order to make the objects, contents and advantages of the present invention clearer, the following detailed description of the embodiments of the present invention will be made in conjunction with the accompanying drawings and examples.

The invention relates to the field of environmental wind field analysis, in particular to a method for calculating multilayer wind direction and wind speed probability of a wind measuring tower through technical research, and provides support for wind energy resource assessment, engineering structure design and shallow wind field forecasting.

In view of the above, the wind direction and wind speed probability density function is obtained by fitting the wind speed with gamma distribution and fitting the wind direction with mixed von mises distribution. And then, wind direction and wind speed combined probability density is obtained by calculating the wind direction and wind speed correlation coefficient and the wind direction and wind speed cumulative probability density. On the basis of wind direction and wind speed probability density data, a probability density graph of a multilayer wind field is drawn through matplotlib, and a probability distribution model of wind direction and wind speed of the multilayer wind field is obtained. The method is suitable for wind energy resource assessment, engineering structure design and shallow wind field forecasting.

In order to achieve the purpose, the invention adopts the following technical scheme: the method for realizing the probability distribution of the multilayer wind direction and the wind speed of the anemometer tower comprises the following steps:

s1, layer-by-layer wind direction and wind speed probability distribution:

and (6) processing data, and abandoning a static wind field.

And fitting the wind speed probability distribution.

And fitting wind direction probability distribution.

S2, wind direction and wind speed joint probability distribution:

and calculating wind direction and wind speed cumulative probability.

And calculating the correlation coefficient of the wind direction and the wind speed.

And calculating the wind direction and wind speed joint probability.

S3, a multilayer wind direction and wind speed probability distribution model:

and (5) processing wind direction and wind speed joint probability data of each layer.

And (4) probability distribution of multilayer wind directions and wind speeds.

The multilayer wind direction and wind speed probability is distributed from the horizontal to the horizontal.

The method specifically comprises the following steps:

s1, layer-by-layer wind direction and wind speed probability distribution:

s11, data processing

And aggregating the minute wind field data hour by hour to respectively obtain the maximum wind speed hour by hour of 10 meters, the maximum wind speed hour by hour of 50 meters and the maximum wind speed hour by hour of 90 meters and the corresponding wind direction. And the wind less than 0.5m/s is defined as a static wind field, and the data of the static wind field is discarded.

S12, fitting each layer of wind speed data by gamma distribution

The probability density function of the gamma distribution is:

where v is the wind speed, η is the shape parameter, and γ is the scale parameter.

And calculating the shape parameter, the position parameter and the scale parameter of gamma distribution according to the maximum likelihood estimation method for the wind speed data of each layer to obtain a corresponding wind speed distribution probability density function. In particular by the scientific computing library scipy of python.

S13, fitting each layer of wind direction data by using mixed von mises distribution

The probability density function of the mixed von mises distribution is as follows:

where θ is the wind direction, i is the number of von mises distribution mixing groups, and w_iDistributing the weight occupied by each component; mu.s_iLocation parameter, k, for von mises distribution_iIs a scale parameter; i is₀(k_i) Is a modified zero order first class Bessel function.

The number of mixing groups is set to 8. The equal intervals of 0-360 degrees are divided into 8 directions, and the thresholds of the directions are (45 degrees, 90 degrees, 135 degrees, 180 degrees, 225 degrees, 270 degrees, 315 degrees and 360 degrees) respectively. And obtaining a probability accumulation value, namely the weight w, of the wind direction data at each azimuth.

And calculating parameters of each group of von mises distribution by adopting a maximum likelihood estimation method. Calculating the average sine value corresponding to each groupAnd cosine valueAnd further obtaining the position parameter mu of each group of distribution.

At the mean sine valueAnd cosine valueOn the basis, a scale parameter k is obtained according to an empirical function.

If the weight w of each group is close to or equal to 0, the group is removed, and the rest groups are subjected to iterative calculation of w, mu and k.

S2, wind direction and wind speed joint probability distribution layer by layer:

s21, calculating wind direction and wind speed cumulative distribution probability

On the basis of the probability density function, the cumulative probability densities F (theta) and F (upsilon) of the wind direction and the wind speed are calculated through integration. The integration step of the wind direction cumulative probability density F (theta) is 1 DEG, and the integration step of the wind speed cumulative probability density F (upsilon) is 0.5 m/s.

S22, calculating the correlation coefficient of wind direction and wind speed

And calculating sine values and cosine values of the wind direction data. The calculation formula of the wind direction and wind speed correlation coefficient is as follows:

wherein r is_vcIs the correlation coefficient, r, of each wind speed with the corresponding wind direction cosine value_vsFor the correlation coefficient, r, of each wind speed with the sine value of the corresponding wind direction_csIs the correlation coefficient between the wind direction cosine value and the sine value. The correlation coefficient of the wind direction and the wind speed is r.

S23, wind direction and wind speed joint probability distribution function

The wind direction and wind speed joint probability distribution function is as follows:

f(ν,θ)＝f(ν)f(θ)[1+δ(1-2F(υ)(1-2F(θ))]

v is the wind speed, and theta is the wind direction; f (nu) is a wind speed probability density function, and f (theta) is a wind direction probability density function; r is a wind direction and wind speed correlation coefficient; f (upsilon) is a wind speed cumulative density function, and F (theta) is a wind direction cumulative density function. And substituting the corresponding elements and functions into a formula to obtain a data set of wind direction and wind speed joint probability distribution.

S3, a multilayer wind direction and wind speed probability distribution model:

s31, processing wind direction and wind speed joint probability data of each layer

And acquiring the maximum value of the wind direction and wind speed joint probability f (v, theta) of each layer of 10 meters, 50 meters and 90 meters, and taking the maximum value as the interval delta h in the Z-axis direction of the multilayer modeling.

S32 probability distribution of multilayer wind direction and wind speed

The wind direction and wind speed joint probability f (v, theta) of each layer of 10 meters, 50 meters and 90 meters is used as input, three-dimensional drawing is carried out by utilizing matplotlib of python, the elevation angle in the visual angle is set to be 10 degrees, the azimuth angle is set to be 210 degrees, the drawing color is set to be 'rainbow', and three-dimensional surface drawing is carried out by utilizing a plot _ surface function. 10 meters are drawn on the first layer, then 50 meters are drawn at intervals of deltah as the second layer, and then a third layer, i.e. 90 meters, is drawn at intervals of deltah. And obtaining the combined probability of the wind direction and the wind speed of the multilayer.

S33, multi-layer wind direction and wind speed probability distribution

And averaging the wind direction and wind speed joint probabilities f (v, theta) of 10 meters, 50 meters and 90 meters to obtain a joint probability average value. Then, the probability average value is subtracted by the joint probabilities of 10 meters, 50 meters and 90 meters respectively to obtain the joint probability interval of each layer. And (3) repeating the step 1) and the step 2), setting the drawing color as 'seismic' (the negative pitch is cool-tone blue, and the positive pitch is warm-tone red), and obtaining the pitch distribution of the wind direction and wind speed joint probability of the multilayer wind field.

The specific embodiment is as follows:

the method for realizing the multilayer wind direction and wind speed probability distribution of the anemometer tower mainly comprises the following steps: fitting to obtain wind direction and wind speed probability distribution; wind direction and wind speed joint probability distribution; and establishing a multilayer wind direction and wind speed probability distribution model.

S1, layer-by-layer wind direction and wind speed probability distribution:

s11, data processing

The height of the anemometer tower is 100 meters, wherein sensors for wind speed and wind direction are arranged at 10 meters, 50 meters and 90 meters, and minute-by-minute wind field data can be obtained through measurement. And (4) aggregating the minute wind field data hour by hour to obtain the maximum wind speed hour by hour of 10 meters, 50 meters and 90 meters and the corresponding wind direction. And the wind less than 0.5m/s is defined as a static wind field, and the data of the static wind field is discarded.

S12, fitting each layer of wind speed data by gamma distribution

The probability density function of the gamma distribution is:

where v is the wind speed, η is the shape parameter, and γ is the scale parameter.

And calculating shape parameters and scale parameters of gamma distribution according to the maximum likelihood estimation method for the wind speed data of each layer to obtain a corresponding wind speed distribution probability density function. The specific implementation is realized by a scientific computing library scipy of python (scipy.

S13, fitting each layer of wind direction data by using mixed von mises distribution

The probability density function of the mixed von mises distribution is:

where θ is the wind direction, i is the number of von mises distribution mixing groups, and w_iDistributing the weight occupied by each component; mu.s_iLocation parameter, k, for von mises distribution_iIs a scale parameter; i is₀(k_i) Is a modified zero order first class Bessel function.

The probability density function of the hybrid von mises distribution is calculated in an iterative manner.

The number of mixing groups can be defined by self, and is not suitable to be too large in consideration of the actual condition of wind direction. The number of mixing groups is here set to 8, depending on the usual orientation of the wind direction. Namely, the directions are divided into 8 directions at equal intervals of 0-360 degrees, and the thresholds of the directions are (45 degrees, 90 degrees, 135 degrees, 180 degrees, 225 degrees, 270 degrees, 315 degrees and 360 degrees) respectively. And calculating the proportion of each azimuth wind direction data sample in the total wind direction sample, and taking the proportion as a probability accumulation value of the wind direction data in each azimuth, namely the weight w.

And calculating parameters of each group of von mises distribution by adopting a maximum likelihood estimation method. Calculating the average sine value corresponding to each group of wind direction dataAnd cosine valueFurther obtain eachThe position parameter μ of the group distribution.

At the mean sine valueAnd cosine valueOn the basis, a scale parameter k is obtained according to an empirical function.

If the weight w of each group is close to or equal to 0, the group is removed, and the rest groups are subjected to iterative calculation of w, mu and k.

S2, wind direction and wind speed joint probability distribution layer by layer:

s21, calculating wind direction and wind speed cumulative distribution probability

On the basis of the probability density function, the cumulative probability densities F (theta) and F (upsilon) of the wind direction and the wind speed are calculated through integration. The integration step of the wind direction cumulative probability density F (theta) is 1 DEG, and the integration step of the wind speed cumulative probability density F (upsilon) is 0.5 m/s.

S22, calculating the correlation coefficient of wind direction and wind speed

And calculating the sine value and the cosine value of all the wind direction data one by one to obtain the sine value and the cosine value of each wind direction data. And then calculating the wind direction and wind speed correlation coefficient.

The calculation formula of the wind direction and wind speed correlation coefficient is as follows:

wherein r is_vcIs the correlation coefficient, r, of each wind speed with the corresponding wind direction cosine value_vsFor each wind speed and the sine value of the corresponding wind directionOf correlation coefficient r_csIs the correlation coefficient between the wind direction cosine value and the sine value. The correlation coefficient of the wind direction and the wind speed is r.

S23, wind direction and wind speed joint probability distribution function

The wind direction and wind speed joint probability distribution function is as follows:

f(ν,θ)＝f(ν)f(θ)[1+r(1-2F(υ)(1-2F(θ))]

v is the wind speed, and theta is the wind direction; f (nu) is a wind speed probability density function, and f (theta) is a wind direction probability density function; r is a wind direction and wind speed correlation coefficient; f (upsilon) is a wind speed cumulative density function, and F (theta) is a wind direction cumulative density function. And substituting the corresponding elements and functions into a formula to obtain a data set of wind direction and wind speed joint probability distribution.

S3, a multilayer wind direction and wind speed probability distribution model:

s31, processing wind direction and wind speed joint probability data of each layer

The maximum value of the wind direction and wind speed joint probability f (v, theta) of each layer of 10 meters, 50 meters and 90 meters is obtained, and the maximum value is used as the display interval delta h of each layer in the Z-axis direction in the multilayer wind direction and wind speed probability distribution modeling, so that the probability distribution model can be conveniently displayed.

S32 probability distribution of multilayer wind direction and wind speed

The wind direction and wind speed joint probability f (v, theta) of each layer of 10 meters, 50 meters and 90 meters is used as input, three-dimensional drawing is carried out by utilizing matplotlib of python, the elevation angle in the visual angle is set to be 10 degrees, the azimuth angle is set to be 210 degrees, the drawing color is set to be 'rainbow', and three-dimensional surface drawing is carried out by utilizing a plot _ surface function. 10 meters are drawn on the first layer, then 50 meters are drawn at intervals of deltah as the second layer, and then a third layer, i.e. 90 meters, is drawn at intervals of deltah. And obtaining the combined probability of the wind direction and the wind speed of the multilayer. The illustration is seen in fig. 1.

S33, multi-layer wind direction and wind speed probability distribution

Averaging the wind direction and wind speed combined probabilities f (v, theta) of three layers of 10 meters, 50 meters and 90 meters to obtain a combined probability average valueThen combining with 10 m, 50 m and 90 m respectivelyAnd subtracting the probability average value from the probability to obtain the joint probability distance of each layer. And repeating the step S32, performing three-dimensional drawing by utilizing matplotlib of python, performing three-dimensional surface drawing by utilizing a plot _ surface function, drawing 10 meters on a first layer, drawing 50 meters at an interval delta h to form a second layer, drawing a third layer at an interval delta h to form 90 meters, setting the drawing color to be 'sesamic' (the negative pitch is cool-tone blue, and the positive pitch is warm-tone red), and obtaining the pitch distribution of the wind direction and wind speed joint probability of the multilayer wind field. The illustration can be seen in figure 2.

The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.