CN104915729A - Method and system for carrying out processing on environment prediction factor data of typhoon intensity - Google Patents

Abstract

The invention, which belongs to the technical field of typhoon intensity processing, especially relates to a method and system for carrying out processing on environment prediction factor data of a typhoon intensity. The method comprises: step a, selecting an environment prediction factor correlated with a typhoon intensity change, and calculating and extracting parameter data correlated with the environment prediction factor; and step b, carrying out standardization processing on the parameter data of the environment prediction factor and carrying out principal component analysis processing on the parameter data after the standardization processing, thereby obtaining new environment prediction factor parameter data. According to the invention, repeatability and inner links of all environment prediction factors are considered fully and are eliminated by using the principal component analysis method, so that the accuracy of the typhoon intensity prediction is improved. The method and system are suitable for sample data processing of various typhoon types.

Description

A kind of method and system for the treatment of bench monsoon intensity environmental forecasting factor data

Technical field

The invention belongs to intensity of typhoon processing technology field, particularly relate to a kind of method and system for the treatment of bench monsoon intensity environmental forecasting factor data.

Background technology

Principal component analysis (PCA) is a kind of method of mathematic(al) manipulation, is most of variable that the less variable of hope goes to explain in original data, becomes to be mutually independent or incoherent variable by variables transformations very high for correlativitys many in our hand.Normally select fewer than original variable number, several new variables of variable in most of data can be explained, i.e. so-called major component, and in order to the composite target of interpretation data.As can be seen here, principal component analysis (PCA) is actually a kind of dimension reduction method.It changes into another given one group of correlated variables by linear transformation and organizes incoherent variable, the order arrangement that these new variablees successively decrease successively according to variance.In mathematic(al) manipulation, keep the population variance of variable constant, make the first variable have maximum variance, be called first principal component, bivariate variance is secondary large, and uncorrelated with the first variable, is called Second principal component.The like, n variable just has n major component.

Principal component analysis (PCA) is one of common method in also meteorological upper multivariable analysis.In meteorological statistics, often to research and analyse various meteorological element field, such as SST fields, ice stadium falls.They are made up of irregular net point mostly.If extract the data of these a certain period of period of history, just forming one group with net point is the time dependent sample of spatial point, principal component analysis (PCA) with this sample for analytic target.Principal component analysis (PCA) and regretional analysis, differential analysis are different, and it is a kind of analytical approach instead of a kind of forecasting procedure.It can be decomposed into spatial function part and the function of time (major component) part time dependent meteorological element field.Spatial function part summarizes the Regional Distribution place of field, and this part is time-independent; Function of time part is then made up of the linear combination of spatial point, is called major component, and several leading of these principal components is occupied the very most of of the population variance of former spatial point (variable).

1994, U.S.'s hurricane strength statistical fluctuation scheme (SHIPS) proposes a kind of Strength Changes forecast model utilizing numerical model to forecast sea basin typhoon in 48 hours, what the program adopted is the sea basin typhoon of 1989-1992 years is that data sample utilizes climatic persistence method and ten overall situation predictor, adopt linear regression thought, all environmental variances are introduced model, all F inspection will be carried out after often introducing an explanatory variable, and one by one t inspection is carried out to the explanatory variable be selected into, when the original explanatory variable introduced becomes no longer significantly due to the introducing of explained later variable, then deleted.With guarantee each introduce new variable before only comprise first active variable in regression equation.This is a process repeatedly, until both do not had significant explanatory variable to be selected into regression equation, also, till inapparent explanatory variable useless is rejected from regression equation, to ensure that the explanatory variable collection finally obtained is optimum, thus the change of intensity of typhoon is estimated.The technical matters that the program exists is: first, do not take into full account for the inner link between all environmental forecasting factors, a large amount of environmental forecasting factors may exist repeatability each other, affects the accuracy of intensity of typhoon prediction; Secondly, the typhoon data sample analysis that what the program adopted is in sea basin, has certain limitation for intensity of typhoon forecast scope.

Summary of the invention

The invention provides a kind of method and system for the treatment of bench monsoon intensity environmental forecasting factor data, be intended to solve existing intensity of typhoon forecasting procedure and do not take into full account inner link between all environmental forecasting factors and repeatability, affect the accuracy of intensity of typhoon prediction, and forecast scope there is certain circumscribed technical matters.

The present invention is achieved in that a kind of method for the treatment of bench monsoon intensity environmental forecasting factor data, comprising:

Step a: choose the environmental forecasting factor relevant to intensity change in typhoon, and calculate the supplemental characteristic relevant with extraction environment predictor;

Step b: the supplemental characteristic of the environmental forecasting factor is carried out standardization, and the supplemental characteristic after standardization is carried out principal component analysis (PCA) process, obtain new environmental forecasting factor parameter data.

The technical scheme that the embodiment of the present invention is taked also comprises: in described step b, describedly the supplemental characteristic of the environmental forecasting factor carried out standardization be specially: take minimax method for normalizing that the supplemental characteristic of all environmental forecasting factors is carried out standardization, concrete formula is:

$Y=(Y_{\max }-Y_{\min })*\frac{X-X_{\min }}{X_{\max }-X_{\min }}+Y_{\min }$

Wherein, Ymax, Ymin are that the normalization of minimax is interval, and be designated as (0,1), the matrix that X is formed for the environmental forecasting factor, Xmax, Xmin are the maximin of each environmental forecasting factor.

The technical scheme that the embodiment of the present invention is taked also comprises: in described step b, describedly supplemental characteristic after standardization carried out principal component analysis (PCA) process specifically also comprise: obtain standardization covariance matrix Z by environmental forecasting factor matrix X, concrete formula is:

$R={\left[r_{\mathrm{ij}}\right]}_p\mathrm{xp}=\frac{Z^{T}Z}{n-1}$

Wherein, $r_{\mathrm{ij}}=\frac{\mathrm{\Σ}{z}_{\mathrm{kj}}\·z_{\mathrm{kj}}}{n-1},i,j=\mathrm{1,2}...,p.$

The technical scheme that the embodiment of the present invention is taked also comprises: in described step b, describedly supplemental characteristic after standardization is carried out principal component analysis (PCA) process specifically also comprise: the characteristic root and the characteristic of correspondence vector thereof that solve correlation matrix R, determine major component and major component number; Specifically comprise: the secular equation solving correlation matrix R | R-λ I _p|=0, obtain p characteristic root, determine major component, press determine that major component number is m, to each λ _j, j=1,2 ... m, solving equations Rb=λ _jb obtains unit character vector

The technical scheme that the embodiment of the present invention is taked also comprises: in described step b, describedly supplemental characteristic after standardization is carried out principal component analysis (PCA) process specifically also comprise: the target variable after standardization is converted to major component, and comprehensive evaluation is carried out to m major component, obtain new environmental forecasting factor parameter data; Concrete formula is:

$U_{\mathrm{ij}}=z_i^T{b}_j^a,j=\mathrm{1,2},...,m$

Wherein, U1 is called first principal component, and U2 is called Second principal component..., Up is called p major component.

The technical scheme that the embodiment of the present invention is taked also comprises: the described environmental forecasting factor comprises: storm potentiality POT, POT ², 200-850hpa vertical wind shear SHR, 500-850hpa vertical wind shear SHR, 12 hours intensity change in typhoon, Julian date, 200hpa temperature, 200hpaU wind, 850hpa vertical vorticity, 500hpa vertical vorticity, vertical wind shear and initial storm latitude product SHR*sin, 200hpa horizontal divergence, initial intensity of typhoon, center of typhoon air pressure, latitude, longitude, storm translational speed SPD, Hai Lu ratio, ocean temperature, 850-700hpa relative humidity, 500-300hpa relative humidity, the shear of 500-850hpa zonal wind, the shear of 200-850hpa zonal wind and water vapor flux.

Another technical scheme that the embodiment of the present invention is taked is: a kind of system for the treatment of bench monsoon intensity environmental forecasting factor data, comprise data extraction module and principal component analysis (PCA) module, described data extraction module for choosing the environmental forecasting factor relevant to intensity change in typhoon, and calculates the supplemental characteristic relevant with extraction environment predictor; Described principal component analysis (PCA) module is used for the supplemental characteristic of the environmental forecasting factor to carry out standardization, and the supplemental characteristic after standardization is carried out principal component analysis (PCA) process, obtains new environmental forecasting factor parameter data.

The technical scheme that the embodiment of the present invention is taked also comprises: described principal component analysis (PCA) module also comprises standardization unit, the supplemental characteristic of the environmental forecasting factor is carried out standardization for taking minimax method for normalizing by described standardization unit, and concrete formula is:

$Y=(Y_{\max }-Y_{\min })*\frac{X-X_{\min }}{X_{\max }-X_{\min }}+Y_{\min }$

Wherein, Ymax, Ymin are that the normalization of minimax is interval, and be designated as (0,1), the matrix that X is formed for the environmental forecasting factor, Xmax, Xmin are the maximin of each environmental forecasting factor.

The technical scheme that the embodiment of the present invention is taked also comprises: described principal component analysis (PCA) module also comprises matrix disposal unit and principal component analysis (PCA) unit;

Described matrix disposal unit is used for obtaining standardization covariance matrix Z by environmental forecasting factor matrix X, and concrete formula is:

$R={\left[r_{\mathrm{ij}}\right]}_p\mathrm{xp}=\frac{Z^{T}Z}{n-1}$

Wherein, $r_{\mathrm{ij}}=\frac{\mathrm{\Σ}{z}_{\mathrm{kj}}\·z_{\mathrm{kj}}}{n-1},i,j=\mathrm{1,2}...,p;$

Described principal component analysis (PCA) unit, for solving characteristic root and the characteristic of correspondence vector thereof of correlation matrix R, determines major component and major component number; Wherein, the secular equation of correlation matrix R is solved | R-λ I _p|=0, obtain p characteristic root, determine major component, press determine that major component number is m, to each λ _j, j=1,2 ... m, solving equations Rb=λ _jb obtains unit character vector

The technical scheme that the embodiment of the present invention is taked also comprises: described principal component analysis (PCA) module also comprises major component converting unit and major component evaluation unit;

Described major component converting unit is used for the target variable after standardization to be converted to major component, and concrete formula is:

$U_{\mathrm{ij}}=z_i^T{b}_j^a,j=\mathrm{1,2},...,m$

U1 is called first principal component, and U2 is called Second principal component..., Up is called p major component;

Described major component evaluation unit is used for carrying out comprehensive evaluation to m major component, obtains new environmental forecasting factor parameter data.

The method and system of the treatment bench monsoon intensity environmental forecasting factor data of the embodiment of the present invention have taken into full account the repeatability and inner link that exist between all environmental forecasting factors, and make use of principal component analysis (PCA) and remove repeatability between each environmental forecasting factor and inner link, thus in conjunction with Linear Regression Forecasting Model, improve the accuracy of intensity of typhoon prediction; In addition, the present invention does not have limitation for the research of typhoon data sample, not only can be applied to the analysis of sea basin typhoon data sample, can be used for the sample data process of the multiple typhoon such as coastal waters typhoon, face, land typhoon yet.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of the method for the treatment bench monsoon intensity environmental forecasting factor data of the embodiment of the present invention;

Fig. 2 is the process flow diagram that the supplemental characteristic to the environmental forecasting factor of the embodiment of the present invention carries out the method for principal component analysis (PCA);

Fig. 3 is the structural representation of the system of the treatment bench monsoon intensity environmental forecasting factor data of the embodiment of the present invention.

Embodiment

In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.

Referring to Fig. 1, is the process flow diagram of the method for the treatment bench monsoon intensity environmental forecasting factor data of the embodiment of the present invention.The method of the treatment bench monsoon intensity environmental forecasting factor data of the embodiment of the present invention comprises the following steps:

Step 100: choose multiple environmental forecasting factor relevant to intensity change in typhoon, and the supplemental characteristic relevant with extracting all environmental forecasting factors is calculated from NCEP (Environmental forecasting centre, National Centers for Environmental Prediction) data;

In step 100, the embodiment of the present invention chooses the variable of 24 environmental forecasting factors as forecast intensity of typhoon model, and these 24 environmental forecasting factors are respectively: 1. storm potentiality POT, 2.POT ², 3.200-850hpa vertical wind shear SHR, 4.500-850hpa vertical wind shear SHR, 5.12 hour intensity change in typhoon, 6. Julian date, 7.200hpa temperature, 8.200hpaU wind, 9.850hpa vertical vorticity, 10.500hpa vertical vorticity, 11. vertical wind shear and initial storm latitude product SHR*sin (lat), 12.200hpa horizontal divergence, 13. initial intensity of typhoon, 14. center of typhoon air pressure, 15. latitudes, 16. longitudes, 17. storm translational speed SPD, 18. Hai Lu ratios, 19. ocean temperatures, 20.850-700hpa relative humidity, 21.500-300hpa relative humidity, 22.500-850hpa zonal wind shear, 23.200-850hpa zonal wind shear, 24. water vapor flux (water vapor flux), and the supplemental characteristic of 24 environmental forecasting factors once within every six hours, is monitored from NCEP extracting data, but, not all parametric variable can extracting directly, such as: water vapor flux is the important environmental forecasting factor affecting intensity change in typhoon, this environmental forecasting factor data is not provided in NCEP, need the acquisition methods understanding this parameter of water vapor flux, calculated the precision of acquired results by more various method, by the computation process of the whole experimental formula of codes implement.In other embodiments of the present invention, also can choose the variable of the environmental forecasting factor as forecast intensity of typhoon model of other quantity or type.

Step 200: the supplemental characteristic of all environmental forecasting factors is carried out standardization, and bring the supplemental characteristic after standardization into principal component model and carry out principal component analysis (PCA) process, obtain new environmental forecasting factor parameter data;

In order to clear description step 200, seeing also Fig. 2, is the process flow diagram that the supplemental characteristic to the environmental forecasting factor of the embodiment of the present invention carries out the method for principal component analysis (PCA).The method that the supplemental characteristic to the environmental forecasting factor of the embodiment of the present invention carries out principal component analysis (PCA) specifically comprises the following steps:

Step 210: take minimax method for normalizing that the supplemental characteristic of all environmental forecasting factors is carried out standardization;

In step 210, because the magnitude difference between each different types of environmental forecasting factor is excessive, for reducing the impact of this species diversity on experimental result, the embodiment of the present invention takes minimax method for normalizing that the supplemental characteristic of all environmental forecasting factors is carried out standardization, the supplemental characteristic of all environmental forecasting factors is supposed all to be normalized into (0,1) this is interval, utilizes minimax normalization formula by its normalization:

$Y=(Y_{\max }-Y_{\min })*\frac{X-X_{\min }}{X_{\max }-X_{\min }}+Y_{\min }$

This formula is designated as maxmin criterion definition, wherein Ymax, Ymin are the normalization interval of minimax, be designated as (0,1), the matrix that X is formed for all environmental forecasting factors, Xmax, Xmin are the maximin of each environmental forecasting factor, can be by simplification of a formula finally:

$Y=\frac{X-X_{\min }}{X_{\max }-X_{\min }}$

Step 220: obtain standardization covariance matrix Z by environmental forecasting factor matrix X, concrete formula is:

$R={\left[r_{\mathrm{ij}}\right]}_p\mathrm{xp}=\frac{Z^{T}Z}{n-1}$

Wherein, $r_{\mathrm{ij}}=\frac{\mathrm{\Σ}{z}_{\mathrm{kj}}\·z_{\mathrm{kj}}}{n-1},i,j=\mathrm{1,2}...,p.$

Step 230: the characteristic root and the characteristic of correspondence vector thereof that solve correlation matrix R, determines major component and major component number;

In step 230, the secular equation of correlation matrix R is solved | R-λ I _p|=0, obtain p characteristic root, determine major component, press determine that major component number is m, make the utilization factor of information reach more than 85%, to each λ _j, j=1,2 ... m, solving equations Rb=λ _jb obtains unit character vector

Step 240: the target variable after standardization is converted to major component, concrete formula is:

$U_{\mathrm{ij}}=z_i^T{b}_j^a,j=\mathrm{1,2},...,m$

Wherein, U1 is called first principal component, and U2 is called Second principal component..., Up is called p major component.

Step 250: carry out comprehensive evaluation to m major component, obtains new environmental forecasting factor parameter data;

In step 250, be weighted summation, obtain final evaluation of estimate to m major component, flexible strategy are the variance contribution ratio of each major component.

Referring to Fig. 3, is the structural representation of the system of the treatment bench monsoon intensity environmental forecasting factor data of the embodiment of the present invention.The system of the treatment bench monsoon intensity environmental forecasting factor data of the embodiment of the present invention comprises data extraction module and principal component analysis (PCA) module; Wherein,

Data extraction module for choosing multiple environmental forecasting factor relevant to intensity change in typhoon, and calculates the supplemental characteristic relevant with extracting all environmental forecasting factors from NCEP data, wherein, the embodiment of the present invention chooses the variable of 24 environmental forecasting factors as forecast intensity of typhoon model, comprising: 1. storm potentiality POT, 2.POT ², 3.200-850hpa vertical wind shear SHR, 4.500-850hpa vertical wind shear SHR, 5.12 hour intensity change in typhoon, 6. Julian date, 7.200hpa temperature, 8.200hpaU wind, 9.850hpa vertical vorticity, 10.500hpa vertical vorticity, 11. vertical wind shear and initial storm latitude product SHR*sin (lat), 12.200hpa horizontal divergence, 13. initial intensity of typhoon, 14. center of typhoon air pressure, 15. latitudes, 16. longitudes, 17. storm translational speed SPD, 18. Hai Lu ratios, 19. ocean temperatures, 20.850-700hpa relative humidity, 21.500-300hpa relative humidity, 22.500-850hpa zonal wind shear, 23.200-850hpa zonal wind shear, 24. water vapor flux (water vapor flux), and the supplemental characteristic of 24 environmental forecasting factors once within every six hours, is monitored from NCEP extracting data, but, not all parametric variable can extracting directly, such as: water vapor flux is the important environmental forecasting factor affecting intensity change in typhoon, this environmental forecasting factor data is not provided in NCEP, need the acquisition methods understanding this parameter of water vapor flux, calculated the precision of acquired results by more various method, by the computation process of the whole experimental formula of codes implement.In other embodiments of the present invention, also can choose the variable of the environmental forecasting factor as forecast intensity of typhoon model of other quantity or type.

Principal component analysis (PCA) module is used for the supplemental characteristic of all environmental forecasting factors to carry out standardization, and brings the supplemental characteristic after standardization into principal component model and carry out principal component analysis (PCA) process, obtains new environmental forecasting factor parameter data; Particularly, principal component analysis (PCA) module comprises standardization unit, matrix disposal unit, principal component analysis (PCA) unit, major component converting unit and major component evaluation unit, wherein:

The supplemental characteristic of all environmental forecasting factors is carried out standardization for taking minimax method for normalizing by standardization unit; Wherein, because the magnitude difference between each different types of environmental forecasting factor is excessive, for reducing the impact of this species diversity on experimental result, the embodiment of the present invention takes minimax method for normalizing that the supplemental characteristic of all environmental forecasting factors is carried out standardization, the supplemental characteristic of all environmental forecasting factors is supposed all to be normalized into (0,1) this is interval, utilizes minimax normalization formula by its normalization:

$Y=(Y_{\max }-Y_{\min })*\frac{X-X_{\min }}{X_{\max }-X_{\min }}+Y_{\min }$

This formula is designated as maxmin criterion definition, wherein Ymax, Ymin are the normalization interval of minimax, be designated as (0,1), the matrix that X is formed for all environmental forecasting factors, Xmax, Xmin are the maximin of each environmental forecasting factor, can be by simplification of a formula finally:

$Y=\frac{X-X_{\min }}{X_{\max }-X_{\min }}$

Matrix disposal unit is used for obtaining standardization covariance matrix Z by environmental forecasting factor matrix X, and concrete formula is:

$R={\left[r_{\mathrm{ij}}\right]}_p\mathrm{xp}=\frac{Z^{T}Z}{n-1}$

Wherein, $r_{\mathrm{ij}}=\frac{\mathrm{\Σ}{z}_{\mathrm{kj}}\·z_{\mathrm{kj}}}{n-1},i,j=\mathrm{1,2}...,p.$

Principal component analysis (PCA) unit, for solving characteristic root and the characteristic of correspondence vector thereof of correlation matrix R, determines major component and major component number; Wherein, the secular equation of correlation matrix R is solved | R-λ I _p|=0, obtain p characteristic root, determine major component, press determine that major component number is m, make the utilization factor of information reach more than 85%, to each λ _j, j=1,2 ... m, solving equations Rb=λ _jb obtains unit character vector

Major component converting unit is used for the target variable after standardization to be converted to major component, and concrete formula is:

$U_{\mathrm{ij}}=z_i^T{b}_j^a,j=\mathrm{1,2},...,m$

U1 is called first principal component, and U2 is called Second principal component..., Up is called p major component.

Major component evaluation unit is used for carrying out comprehensive evaluation to m major component, obtains new environmental forecasting factor parameter data; Wherein, be weighted summation, obtain final evaluation of estimate to m major component, flexible strategy are the variance contribution ratio of each major component.

The method and system of the treatment bench monsoon intensity environmental forecasting factor data of the embodiment of the present invention have taken into full account the repeatability and inner link that exist between all environmental forecasting factors, and make use of principal component analysis (PCA) and remove repeatability between each environmental forecasting factor and inner link, thus in conjunction with Linear Regression Forecasting Model, improve the accuracy of intensity of typhoon prediction; In addition, the present invention does not have limitation for the research of typhoon data sample, not only can be applied to the analysis of sea basin typhoon data sample, can be used for the sample data process of the multiple typhoon such as coastal waters typhoon, face, land typhoon yet.

The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.

Claims (10)

1. a method for treatment bench monsoon intensity environmental forecasting factor data, comprising:

Step a: choose the environmental forecasting factor relevant to intensity change in typhoon, and calculate the supplemental characteristic relevant with extraction environment predictor;

Step b: the supplemental characteristic of the environmental forecasting factor is carried out standardization, and the supplemental characteristic after standardization is carried out principal component analysis (PCA) process, obtain new environmental forecasting factor parameter data.

2. the method for the treatment of bench monsoon intensity environmental forecasting factor data according to claim 1, it is characterized in that, in described step b, describedly the supplemental characteristic of the environmental forecasting factor carried out standardization be specially: take minimax method for normalizing that the supplemental characteristic of all environmental forecasting factors is carried out standardization, concrete formula is:

$(Y_{\max }-Y_{\min })*\frac{X-X_{\min }}{X_{\max }-X_{\min }}{Y}_{\min },$

Wherein, Ymax, Ymin are that the normalization of minimax is interval, and be designated as (0,1), the matrix that X is formed for the environmental forecasting factor, Xmax, Xmin are the maximin of each environmental forecasting factor.

3. the method for the treatment of bench monsoon intensity environmental forecasting factor data according to claim 1, it is characterized in that, in described step b, describedly supplemental characteristic after standardization carried out principal component analysis (PCA) process specifically also comprise: obtain standardization covariance matrix Z by environmental forecasting factor matrix X, concrete formula is:

$R={\left[r_{\mathrm{ij}}\right]}_p\mathrm{xp}=\frac{Z^{T}Z}{n-1},$

Wherein, $r_{\mathrm{ij}}=\frac{\mathrm{\Σ}{z}_{\mathrm{kj}}\·z_{\mathrm{kj}}}{n-1},i,j=\mathrm{1,2}...,p.$

4. the method for the treatment of bench monsoon intensity environmental forecasting factor data according to claim 3, it is characterized in that, in described step b, describedly supplemental characteristic after standardization is carried out principal component analysis (PCA) process specifically also comprise: the characteristic root and the characteristic of correspondence vector thereof that solve correlation matrix R, determine major component and major component number; Specifically comprise: the secular equation solving correlation matrix R | R-λ I _p|=0, obtain p characteristic root, determine major component, press determine that major component number is m, to each λ _j, j=1,2 ... m, solving equations Rb=λ _jb obtains unit character vector

5. the method for the treatment of bench monsoon intensity environmental forecasting factor data according to claim 4, it is characterized in that, in described step b, describedly supplemental characteristic after standardization is carried out principal component analysis (PCA) process specifically also comprise: the target variable after standardization is converted to major component, and comprehensive evaluation is carried out to m major component, obtain new environmental forecasting factor parameter data; Concrete formula is:

$U_{\mathrm{ij}}=z_i^T{b}_j^a,j=\mathrm{1,2},...,m,$

Wherein, U1 is called first principal component, and U2 is called Second principal component..., Up is called p major component.

6. the method for the treatment bench monsoon intensity environmental forecasting factor data according to any one of claim 1 to 5, it is characterized in that, the described environmental forecasting factor comprises: storm potentiality POT, POT ², 200-850hpa vertical wind shear SHR, 500-850hpa vertical wind shear SHR, 12 hours intensity change in typhoon, Julian date, 200hpa temperature, 200hpaU wind, 850hpa vertical vorticity, 500hpa vertical vorticity, vertical wind shear and initial storm latitude product SHR*sin, 200hpa horizontal divergence, initial intensity of typhoon, center of typhoon air pressure, latitude, longitude, storm translational speed SPD, Hai Lu ratio, ocean temperature, 850-700hpa relative humidity, 500-300hpa relative humidity, the shear of 500-850hpa zonal wind, the shear of 200-850hpa zonal wind and water vapor flux.

7. the system of a treatment bench monsoon intensity environmental forecasting factor data, it is characterized in that, comprise data extraction module and principal component analysis (PCA) module, described data extraction module for choosing the environmental forecasting factor relevant to intensity change in typhoon, and calculates the supplemental characteristic relevant with extraction environment predictor; Described principal component analysis (PCA) module is used for the supplemental characteristic of the environmental forecasting factor to carry out standardization, and the supplemental characteristic after standardization is carried out principal component analysis (PCA) process, obtains new environmental forecasting factor parameter data.

8. the system for the treatment of bench monsoon intensity environmental forecasting factor data according to claim 7, it is characterized in that, described principal component analysis (PCA) module also comprises standardization unit, the supplemental characteristic of the environmental forecasting factor is carried out standardization for taking minimax method for normalizing by described standardization unit, and concrete formula is:

$(Y_{\max }-Y_{\min })*\frac{X-X_{\min }}{X_{\max }-X_{\min }}{Y}_{\min },$

Wherein, Ymax, Ymin are that the normalization of minimax is interval, and be designated as (0,1), the matrix that X is formed for the environmental forecasting factor, Xmax, Xmin are the maximin of each environmental forecasting factor.

9. the system for the treatment of bench monsoon intensity environmental forecasting factor data according to claim 8, is characterized in that: described principal component analysis (PCA) module also comprises matrix disposal unit and principal component analysis (PCA) unit;

Described matrix disposal unit is used for obtaining standardization covariance matrix Z by environmental forecasting factor matrix X, and concrete formula is:

$R={\left[r_{\mathrm{ij}}\right]}_p\mathrm{xp}=\frac{Z^{T}Z}{n-1},$

Wherein, $r_{\mathrm{ij}}=\frac{\mathrm{\Σ}{z}_{\mathrm{kj}}\·z_{\mathrm{kj}}}{n-1},i,j=\mathrm{1,2}...,p;$

Described principal component analysis (PCA) unit, for solving characteristic root and the characteristic of correspondence vector thereof of correlation matrix R, determines major component and major component number; Wherein, the secular equation of correlation matrix R is solved | R-λ I _p|=0, obtain p characteristic root, determine major component, press determine that major component number is m, to each λ _j, j=1,2 ... m, solving equations Rb=λ _jb obtains unit character vector

10. the system for the treatment of bench monsoon intensity environmental forecasting factor data according to claim 9, is characterized in that: described principal component analysis (PCA) module also comprises major component converting unit and major component evaluation unit;

Described major component converting unit is used for the target variable after standardization to be converted to major component, and concrete formula is:

$U_{\mathrm{ij}}=z_i^T{b}_j^a,j=\mathrm{1,2},...,m,$

U1 is called first principal component, and U2 is called Second principal component..., Up is called p major component;

Described major component evaluation unit is used for carrying out comprehensive evaluation to m major component, obtains new environmental forecasting factor parameter data.