CN103150482A - Method for determining parameter influence factors of cut tobacco drying process based on partial least square (PLS) - Google Patents

Abstract

The invention discloses a method for determining parameter influence factors of a cut tobacco drying process based on a partial least square (PLS). The method comprises the following steps of: selecting target variables and independent variables of process data in the cut tobacco drying process, and acquiring the historical data of the target variables and the independent variables; screening the historical data by adopting a Euclidean distance method, and rejecting abnormal data from the historical data to obtain sample data; and establishing a PLS model of the independent variables x and the target variables y by adopting the PLS, and calculating the magnitudes of influence factors of the independent variables x which influence the target variables y according to the model. The method is simple and easy to implement; and a manual experience-based judgment method is replaced by the method, so that the magnitudes of the influence factors can be obtained quickly and accurately.

Description

A kind of method of determining the baking silk effects of process parameters factor based on PLS

Technical field

The present invention relates to a kind of method of determining the baking silk effects of process parameters factor based on PLS.

Background technology

Baking tabacco scrap technique is requisite link in the Cigarette processing during Primary Processing, and baking silk technological process is larger on physical qualities and aesthetic quality's impact of cigarette product.The parameter of impact baking silk technique has more than 10, and present stage can be adjusted state parameter on supervisory control comuter.Extremely yet (the device parameter detected value occurs abnormal when producing appearance, supplied materials offset setting etc.), can't be according to effect of parameters factor size in baking silk technique, find accurately, rapidly the root of abnormal situation, and it is whole to do the wage adjustment of conducting oneself of equivalent compensation and technique experience by automatic control system.Therefore, be badly in need of to propose a kind of method and determine the size to other effect of parameters factors in whole production run of each parameter in baking silk technique.

Summary of the invention

The invention provides a kind of method of determining the baking silk effects of process parameters factor based on PLS, its purpose is, overcomes and can't determine to dry by the fire a size of the effects of process parameters factor in prior art.

The technical solution used in the present invention is as follows:

A kind of method of determining the baking silk technogenic influence factor based on PLS comprises the following steps:

Step 1: the independent variable x and the target variable y that set baking silk technique, independent variable x chooses from HT vapor pressure, HT steam flow, entrance moisture, cylinder entrance hot blast temperature, cylinder entrance hot blast air quantity, barrel temperature, barrel vapor pressure, humidity discharging throttle opening and discharging cover negative pressure, target variable y chooses from discharging baking silk water percentage, discharging baking silk temperature and humidity discharging humidity, gather the historical data of independent variable x and target variable y, wherein the proportion of historical data is T;

Step 2: adopt Euclidean distance method that historical data is screened, reject the abnormal data in historical data, obtain n group sample data, sample data consists of target variable matrix Y _njWith independent variable matrix X _ni, target variable Y has j, and independent variable X has i;

The baking silk process data that gathers when described abnormal data refers to that drying by the fire a process equipment breaks down;

Calculate between the independent variable x and independent variable x of baking silk technique, between target variable y and target variable y and the correlation coefficient matrix between independent variable x and target variable y;

We find to have serious multiple correlation between each independent variable in baking silk technique by this correlation matrix, show simultaneously between independent variable x and target variable y to have simple linear relationship, we determine to select the PLS model to analyze independent variable to the relation that affects of target variable under this condition.

Step 3: the sample data after the screening that obtains according to step 2, set up independent variable X _niWith target variable Y _njPartial least square method PLS model, calculate the size of the factor of influence of each the independent variable x that affects each target variable y according to the PLS model, its specific operation process is as follows:

1) utilize partial least square method PLS and cross validation test method to determine that the PLS component number in sample data is h;

Its difference of PLS and principal component analysis (PCA) (PCA) also is used for describing variable X when being for the description variable Y factor.In order to realize this point, be to go the factor of compute matrix X with the row of matrix Y on mathematics, meanwhile, the factor of matrix Y is gone prediction by the row of matrix X, in the hope of take into account X and Y all match get reasonable load P LS composition.

2) iterative algorithm in the employing partial least square method, be obtained from matrix of variables X _niWeight matrix W, objective matrix Y _njLoad factor matrix Q ^TWith independent variable matrix X _niLoading matrix P ^TIndependent variable matrix X _niWeight matrix W, objective matrix Y _njLoad factor matrix Q ^TWith independent variable matrix X _niLoading matrix P ^TBe calculated as prior art, weight matrix W is the capable i row of h, load factor matrix Q ^TThe capable j row of h, loading matrix P ^TThe capable i row of h;

$W=\left[\begin{array}{cccc}{w}_{11}& ...& & w_{1i}\\ w_{21}& ...& & w_{2i}\\ ...& ...& & ...\\ ...& ...& & ...\\ w_{h1}& ...& & w_{\mathrm{hi}}\end{array}\right]$

$Q^T=\left[\begin{array}{cccc}{a}_{11}& ...& & a_{1j}\\ a_{21}& ...& & a_{2j}\\ ...& ...& & ...\\ ...& ...& & ...\\ a_{h1}& ...& & a_{\mathrm{hj}}\end{array}\right],P^T=\left[\begin{array}{cccc}{c}_{11}& ...& & c_{1i}\\ c_{21}& ...& & c_{2i}\\ ...& ...& & ...\\ ...& ...& & ...\\ c_{h1}& ...& & c_{\mathrm{hi}}\end{array}\right]$

3) determine independent variable matrix X resolute t _hWith objective matrix Y resolute u _h:

Independent variable matrix X's must resolute t _hFor:

t _h=c _h1x ₁+c _h2x ₂+…+c _hix _i

Objective matrix Y's must resolute u _hFor:

u _h=a _h1y ₁+…+a _hjy _j

Wherein, h is the component number of sample data, and i is the independent variable number, and j is the target variable number;

4) the weight matrix W of independent variable matrix X carried out normalization calculating, obtaining two inside correlation matrixes that get between resolute is V:

$V=\left[\begin{array}{cccc}{v}_1& 0& ...& 0\\ 0& ...& ...& ...\\ ...& ...& ...& ...\\ ...& ...& ...& 0\\ 0& ...& 0& v_h\end{array}\right]$

Thereby two internal relations formulas that get between resolute are expressed as:

u ₁=v ₁t ₁+ e, u ₂=v ₂t ₂+ e, u ₃=v ₃t ₃+ e ... u _h=v _ht _h+ e, e are regression residuals;

5) according to B=V (P ^TV) ^-1Q ^TCalculate the matrix of coefficients B in Y=XB:

$B=\left[\begin{array}{cccc}{b}_{11}& ...& & b_{1i}\\ b_{21}& ...& & b_{2i}\\ ...& ...& & ...\\ ...& ...& & ...\\ b_{j1}& ...& & b_{\mathrm{ji}}\end{array}\right]$

Thereby the relation that finally obtains between independent variable x and target variable y is as follows:

y ₁=b ₁₁x ₁+b ₁₂x ₂+b ₁₃x ₃+……+b _1ix _i

y ₂=b ₂₁x ₁+b ₂₂x ₂+b ₂₃x ₃+……+b _2ix _i

……

y _j=b _j1x ₁+b _j2x ₂+b _j3x ₃+……+b _jix _i

Step 4: extract independent variable coefficient absolute value in above-mentioned formula greater than the independent variable of the 1/i factor of influence as corresponding target variable, i is the number of independent variable X.

If the absolute value of independent variable coefficient is greater than 1/i, namely can be considered this independent variable target variable is had certain impact, if differing by more than, the absolute value of the coefficient that the coefficient of absolute value maximum and absolute value are second largest equals 1/i, independent variable that can clear and definite Coefficient of determination absolute value maximum is the main affecting factors of target variable, otherwise just the independent variable of these two larger coefficients is defined as main affecting factors.

Beneficial effect

The invention provides a kind of method of determining the baking silk effects of process parameters factor based on PLS, by target variable and the independent variable in selected baking silk technological process, gather the historical data of target variable and independent variable; Adopt Euclidean distance method that historical data is screened, reject the abnormal data in historical data, obtain sample data; Calculate the independent variable x of baking silk technique and the correlation coefficient matrix between independent variable x, target variable y and target variable y and independent variable x and target variable y; Observe the correlativity size between independent variable and target variable, adopt least square method PLS to set up the PLS model of independent variable x and target variable y, calculate the size of the factor of influence of the independent variable x that obtains to affect each target variable y according to model.The method is simple, be easy to realize, uses this method and substitutes the artificial experience judgement, obtains fast and accurately the size of factor of influence.

Description of drawings

Fig. 1 is schematic flow sheet of the present invention;

Fig. 2 is that each independent variable is to the histogram of the factor of influence size of target variable y1;

Fig. 3 is that each independent variable is to the histogram of the factor of influence size of target variable y2.

Embodiment

The present invention will be further described below in conjunction with accompanying drawing.

Factory produces certain model cigarette as example take certain cigarette, an a kind of method of determining the baking silk effects of process parameters factor based on PLS, and its schematic flow sheet is as shown in Figure 1.

1) target variable y and independent variable x determine to reach the collection of historical data

Technological parameter involved in baking silk technique is a lot, comprises the process data (HT vapor pressure, HT steam flow, inlet water grade) of HT section; Cylinder entrance process data (cylinder entrance hot blast temperature, cylinder entrance hot blast air quantity etc.); Barrel supplemental characteristic (barrel temperature, barrel vapor pressure etc.); Humidity discharging wind supplemental characteristic (humidity discharging throttle opening, discharging cover negative pressure etc.) and pipe tobacco discharging supplemental characteristic (discharging moisture content of cut tobaccos, discharging pipe tobacco temperature etc.).The present embodiment is selected take the water percentage of discharging pipe tobacco and outlet temperature as target variable (y ₁, y ₂), since HT entrance moisture, HT vapor pressure, HT steam flow, hot blast temperature, hot blast air door, discharging cover pressure, cylinder wall temperature, storage heater steam flow, pipe tobacco instantaneous delivery be independent variable (x ₁, x ₂... x ₉);

(sample data of 2012-3-01～2012-4-25), the sample frequency of data is set to 60s to extract 2 months from historical data base.These data comprise target variable discharging moisture content of cut tobaccos and outlet temperature, form original object variable sample Y; Extract simultaneously whole argument datas, the original independent variable sample of these data formations X;

2) screening sample

Owing to may having the abnormal occurrencies such as equipment failure, shutdown in historical production run, the sample of improper production may appear in the sample data of extraction, and the existence of these samples can affect the final analysis result of PLS, therefore must screen sample.This programme adopts Euclidean distance method (Euclidean distance) deletion outliers, through the total n group of the data after the screening sample Variable Selection, independent variable sample X _nj(i=1,2 ..., n; J=1,2 ..., 9), target variable sample Y _nj(i=1,2 ..., n; J=1,2).

3) correlation coefficient matrix calculates

Calculate the independent variable x of baking silk technique and the correlation coefficient matrix between independent variable x, target variable y and target variable y and independent variable x and target variable y.

Related coefficient claims again linearly dependent coefficient, and it is the index of weighing linear dependence degree between variable, and sample correlation coefficient represents with r, and the Calculation of correlation factor formula is as follows:

$r=\frac{\mathrm{\Σ}(X-\stackrel{\‾}{X})(Y-\stackrel{\‾}{Y})}{\sqrt{\mathrm{\Σ}{(X-\stackrel{\‾}{X})}^2\mathrm{\Σ}{(Y-\stackrel{\‾}{Y})}^2}}=\frac{\mathrm{\ΣXY}-\frac{\mathrm{\ΣX}\·\mathrm{\ΣY}}{n}}{\sqrt{[\mathrm{\Σ}{X}^2-\frac{{\left(\mathrm{\ΣX}\right)}^2}{n}][\mathrm{\Σ}{Y}^2-\frac{{\left(\mathrm{\ΣY}\right)}^2}{n}]}}$

Table 1: baking silk process parameter table

As shown in table 2 based on related coefficient between each variable of the sample data calculating after step 2 screening, correlation analysis matrix between table 2 expression baking silk technological process important parameter, numerical value in form represents the correlativity of two parameters in the gauge outfit of corresponding row and column, negative number representation negative correlation (namely another parameter value corresponding to parameter value increase can reduce), in table, gray cell lattice institutes column data absolute value greater than 50, can find out that therefrom between corresponding two parameters, correlativity is stronger.

Table 2: baking silk technological parameter correlation matrix (%)

4) carry out the PLS modeling

Utilize partial least square method to carry out computing to data, by the cross validation test method, determine that the PLS component number is 5.Through the iterative algorithm in partial least square method, obtain the load factor matrix Q of objective matrix Y ^TLoading matrix P with technological parameter matrix X ^T:

$W=\left[\begin{array}{ccccccccc}0.47& -0.17& -0.77& -0.22& -0.03& 0.27& -0.06& -0.11& 0.15\\ -0.12& 0.14& -0.01& 0.03& -0.54& -0.16& -0.57& -0.56& 0.05\\ 0.13& -0.24& -0.2& 0.31& -0.06& -0.59& -0.38& 0.54& 0.05\\ -0.05& 0.7& -0.03& -0.18& -0.12& 0.22& -0.2& 0.47& 0.39\\ -0.1& -0.11& 0.01& -0.03& 0& 0.5& -0.47& 0.27& -0.66\end{array}\right]$

$Q^T=\left[\begin{array}{cc}0.47& -0.88\\ -0.04& -1\\ 0.92& -0.4\\ 0.62& 0.79\\ 0.98& -0.19\end{array}\right]$

$P^T=\left[\begin{array}{ccccccccc}63.05& -28.29& -87.8& -32.29& 34.62& 51.18& 36.97& 13.82& 12.21\\ -23.73& 58.02& 1.67& -9.28& -91.71& -4.22& -88.03& -86.37& 25.69\\ 16.07& -25.3& -17.68& 30.34& -4.97& -65.25& -16.64& 33.15& 10.25\\ -8.9& 54.64& -0.39& -18.41& -5.66& 13.42& -14.43& 35.98& 56.33\\ -9.99& -7.59& -3.33& -3.73& -16.24& 40.41& -28.7& 27.53& -64.56\end{array}\right]$

Technological parameter matrix X's must resolute t ₁-t ₅Just can calculate by formula following according to formula:

t ₁=63.05x ₁-28.29x ₂-87.8x ₃-32.29x ₄+34.62x ₅+51.18x ₆+36.97x ₇+13.82x ₈+12.21x ₉

t ₂=-23.73x ₁+58.02x ₂+1.67x ₃-9.28x ₄-91.71x ₅-4.22x ₆-88.03x ₇-86.37x ₈+25.69x ₉

t ₃=16.07x ₁-25.3x ₂-17.68x ₃+30.34x ₄-4.97x ₅-65.25x ₆-16.64x ₇+33.15x ₈+10.25x ₉

t ₄=-8.9x ₁+54.64x ₂-0.39x ₃-18.41x ₄-5.66x ₅+13.42x ₆-14.43x ₇+35.98x ₈+56.33x ₉

t ₅=-9.99x ₁-7.59x ₂-3.33x ₃-3.73x ₄-16.24x ₅+40.41x ₆-28.7x ₇+27.53x ₈-64.56x ₉

Equally, objective matrix Y's must resolute u ₁-u ₅For:

u ₁=0.47y ₁-0.88y ₂

u ₂=-0.04y ₁-y ₂

u ₃=0.92y ₁-0.4y2

u ₄=0.62y ₁+0.79y ₂

u ₅=0.98y1-0.19y ₂

Two inside correlation matrixes that get between resolute are:

$V=\left[\begin{array}{ccccc}70.61& 0& 0& 0& 0\\ 0& 33.63& 0& 0& 0\\ 0& 0& 27.18& 0& 0\\ 0& 0& 0& 17.39& 0\\ 0& 0& 0& 0& 8.33\end{array}\right]$

Thereby two internal relations formulas that get between resolute are expressed as:

u ₁=70.61t ₁+e?u ₂=33.63t ₂+e?u ₃=27.18t ₃+e?u ₄=17.39t ₄+e?u ₅=8.33t5+e

E is regression residuals.

Finally can calculate the B that concerns of target vector and technological parameter vector:

$B=\left[\begin{array}{ccccccccc}0.18& -0.04& -0.33& 0& -0.09& -0.04& -0.19& 0.24& 0.05\\ -0.33& 0.16& 0.54& 0.07& 0.17& -0.04& 0.17& 0.23& -0.02\end{array}\right]$

y ₁=0.18x ₁-0.04x ₂-0.33x ₃+0x ₄-0.09x ₅-0.04x ₆-0.19x ₇+0.24x ₈+0.05x ₉

y2=-0.33x ₁+0.16x ₂+0.54x ₃+0.07x ₄+0.17x ₅-0.04x ₆+0.17x ₇+0.23x ₈-0.02x ₉

The result of calculation that obtains after modeling as shown in Figures 2 and 3, Fig. 2 and Fig. 3 have represented respectively through the related coefficient histogram between resulting moisture content of outlet and outlet temperature and factor of influence after the PLS analysis modeling, can recognize quicklook that from figure those related coefficient absolute values are the principal elements that affects target variable greater than 0.2 factor of influence, can see that also each influence factor is to the influence degree of target variable simultaneously.

Utilize PLS to the modeling of this section historical data, the principal element that has obtained affecting the moisture content of outlet mass property is followed successively by: HT section steam flow, hot blast air quantity, barrel temperature, HT section entrance moisture, hot blast temperature; The principal element that affects the outlet temperature mass property is followed successively by: HT section steam flow, HT section entrance moisture, hot blast air quantity, hot blast temperature, barrel temperature, HT section vapor pressure, discharging cover pressure.

After analyzing through PLS, obtain analysis result, then the technique experience with technologist's reality contrasts.The conclusion that the PLS analysis modeling obtains is substantially identical with the actual condition of production.In sum, this patent can confirm to affect the mass property factor of baking silk, reliable results fast and accurately.

Claims (1)

1. a method of determining the baking silk technogenic influence factor based on PLS, is characterized in that, comprises the following steps:

Step 1: the independent variable x and the target variable y that set baking silk technique, independent variable x chooses from HT vapor pressure, HT steam flow, entrance moisture, cylinder entrance hot blast temperature, cylinder entrance hot blast air quantity, barrel temperature, barrel vapor pressure, humidity discharging throttle opening and discharging cover negative pressure, target variable y chooses from discharging baking silk water percentage, discharging baking silk temperature and humidity discharging humidity, gather the historical data of independent variable x and target variable y, wherein the proportion of historical data is T;

Step 2: adopt Euclidean distance method that historical data is screened, reject the abnormal data in historical data, obtain n group sample data, sample data consists of target variable matrix Y _njWith independent variable matrix X _ni, target variable Y has j, and independent variable X has i;

The baking silk process data that gathers when described abnormal data refers to that drying by the fire a process equipment breaks down;

Step 3: the sample data after the screening that obtains according to step 2, set up independent variable X _{N * i}With target variable Y _{N * j}Partial least square method PLS model, calculate the size of the factor of influence of each the independent variable x that affects each target variable y according to the PLS model, its specific operation process is as follows:

1) utilize partial least square method PLS and cross validation test method to determine that the PLS component number in sample data is h;

2) iterative algorithm in the employing partial least square method, be obtained from matrix of variables X _niWeight matrix W, objective matrix Y _njLoad factor matrix Q ^TWith independent variable matrix X _niLoading matrix P ^T:

$W=\left[\begin{array}{cccc}{w}_{11}& ...& & w_{1i}\\ w_{21}& ...& & w_{2i}\\ ...& ...& & ...\\ ...& ...& & ...\\ w_{h1}& ...& & w_{\mathrm{hi}}\end{array}\right]$

$Q^T=\left[\begin{array}{cccc}{a}_{11}& ...& & a_{1j}\\ a_{21}& ...& & a_{2j}\\ ...& ...& & ...\\ ...& ...& & ...\\ a_{h1}& ...& & a_{\mathrm{hj}}\end{array}\right],P^T=\left[\begin{array}{cccc}{c}_{11}& ...& & c_{1i}\\ c_{21}& ...& & c_{2i}\\ ...& ...& & ...\\ ...& ...& & ...\\ c_{h1}& ...& & c_{\mathrm{hi}}\end{array}\right]$

3) determine independent variable matrix X resolute t _hWith objective matrix Y resolute u _h:

Independent variable matrix X _{N * i}Resolute t _hFor:

t _h=c _h1x ₁+c _h2x ₂+…+c _hix _i

Objective matrix Y _{N * j}Resolute u _hFor:

u _h=a _h1y ₁+…+a _hjy _j

Wherein, h is the component number of sample data, and i is the independent variable number, and j is the target variable number;

4) to independent variable matrix X _{N * i}Weight matrix W carry out normalization and calculate, obtain two the inside correlation matrixes between resolute are V:

$V=\left[\begin{array}{cccc}{v}_1& 0& ...& 0\\ 0& ...& ...& ...\\ ...& ...& ...& ...\\ ...& ...& ...& 0\\ 0& ...& 0& v_h\end{array}\right]$

Thereby two internal relations formulas that get between resolute are expressed as:

u ₁=v ₁t ₁+ e, u ₂=v ₂t ₂+ e, u ₃=v ₃t ₃+ e ... u _h=v _ht _h+ e, e are regression residuals;

5) according to B=V (P ^TV) ^-1Q ^TCalculate the matrix of coefficients B in Y=XB:

$B=\left[\begin{array}{cccc}{b}_{11}& ...& & b_{1i}\\ b_{21}& ...& & b_{2i}\\ ...& ...& & ...\\ ...& ...& & ...\\ b_{j1}& ...& & b_{\mathrm{ji}}\end{array}\right]$

Thereby the relation that finally obtains between independent variable x and target variable y is as follows:

y ₁=b ₁₁x ₁+b ₁₂x ₂+b ₁₃x ₃+……+b _1ix _i

y ₂=b ₂₁x ₁+b ₂₂x ₂+b ₂₃x ₃+……+b _2ix _i

……

y _j=b _j1x ₁+b _j2x ₂+b _j3x ₃+……+b _jix _i

Step 4: extract independent variable coefficient absolute value in above-mentioned formula greater than the independent variable of the 1/i factor of influence as corresponding target variable, i is the number of independent variable X.

Images