CN103675005B - The industrial melt index soft measurement instrument of optimum FUZZY NETWORK and method - Google Patents

Abstract

The invention discloses a kind of industrial melt index soft measurement instrument and method of optimum FUZZY NETWORK.The method is optimized original fuzzy neural network by introducing support vector machine, solves the problem of difficult parameters setting in fuzzy neural network building process.In the present invention, field intelligent instrument, control station are connected with DCS database, hard measurement value display instrument comprises the industrial melt index soft measurement model of optimum FUZZY NETWORK, DCS database is connected with the input end of soft-sensing model, and the output terminal of the industrial melt index soft measurement model of described optimum FUZZY NETWORK is connected with melt index flexible measured value display instrument.Finally, the present invention has on-line measurement, computing velocity is fast, noise resisting ability is strong, promote the good feature of performance.

Description

The industrial melt index soft measurement instrument of optimum FUZZY NETWORK and method

Technical field

The present invention relates to soft measuring instrument and method, particularly relate to a kind of industrial melt index soft measurement instrument and method of optimum FUZZY NETWORK.

Background technology

Polypropylene is a kind of hemicrystalline thermoplastics by propylene polymerization, and have higher resistance to impact, engineering properties is tough, and anti-multiple organic solvent and acid and alkali corrosion, being widely used in industry member, is one of usual modal macromolecular material.Melting index (MI) determines one of important quality index of the final products trade mark during polypropylene is produced, and which determines the different purposes of product.Melting index accurate, measure timely, to production and scientific research, have very important effect and directive significance.But the on-line analysis of melting index is measured and is still difficult to accomplish at present, the in-line analyzer lacking melting index is a subject matter of restriction polypropylene product quality.MI can only pass through hand sampling, off-line assay obtains, and general every 2-4 hour analyzes once, and time lag is large, is difficult to meet the requirement of producing and controlling in real time.

Research work major part in recent years about the online forecasting of MI all concentrates on above artificial neural network, achieves good effect.But artificial neural network also has the shortcoming of himself, the interstitial content of such as over-fitting, hidden layer and parameter is bad determines.Secondly, the DCS data that industry spot collects also because noise, manual operation error etc. are with certain uncertain error, so use the general Generalization Ability of forecasting model of the artificial neural network that determinacy is strong or not.

Nineteen sixty-five U.S. mathematician L.Zadeh first proposed the concept of Fuzzy set.Subsequently fuzzy logic with its problem closer to daily people and the meaning of one's words statement mode, start the classical logic replacing adhering to that all things can represent with binary item.1987, Bart Kosko took the lead in fuzzy theory and neural network to organically combine to have carried out comparatively systematic research.In time after this, the theory of fuzzy neural network and application thereof obtain development at full speed, the proposition of various new fuzzy neural network model and the research of learning algorithm adapted thereof not only accelerate the perfect of fuzzy neural theory, and have also been obtained application widely in practice.

Support vector machine, introduced in 1998 by Vapnik, by structural risk minimization in Using statistics theory study but not general experience structure Method for minimization, original optimal classification surface problem is converted into the optimization problem of its antithesis, thus there is good Generalization Ability, be widely used in pattern-recognition, matching and classification problem.In this programme, support vector machine is used to the linear dimensions in Optimization of Fuzzy neural network model.

Summary of the invention

In order to the measuring accuracy overcoming existing propylene polymerization production process is not high, low to noise sensitivity, promote the deficiency of poor performance, the invention provides a kind of on-line measurement, computing velocity be fast, model upgrades automatically, noise resisting ability is strong, promote industrial melt index soft measurement instrument and the method for the good optimum FUZZY NETWORK of performance.

A kind of industrial melt index soft measurement instrument of optimum FUZZY NETWORK, comprise propylene polymerization production process, for measuring the field intelligent instrument easily surveying variable, for measuring the control station of performance variable, the DCS database of store data and melt index flexible measured value display instrument, described field intelligent instrument, control station is connected with propylene polymerization production process, described field intelligent instrument, control station is connected with DCS database, described soft measuring instrument also comprises the industrial melt index soft measurement model of optimum FUZZY NETWORK, described DCS database is connected with the input end of the industrial melt index soft measurement model of described optimum FUZZY NETWORK, the output terminal of the industrial melt index soft measurement model of described optimum FUZZY NETWORK is connected with melt index flexible measured value display instrument, the industrial melt index soft measurement model of described optimum FUZZY NETWORK comprises:

Data preprocessing module, carries out pre-service for the model training sample will inputted from DCS database, and make the average of training sample be 0, variance is 1, and this process adopts following formula process:

Computation of mean values: $\stackrel{\‾}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{TX}}_i---\left(1\right)$

Calculate variance: ${\mathrm{\σ}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{TX}}_i-\stackrel{\‾}{\mathrm{TX}})---\left(2\right)$

Standardization: $X=\frac{\mathrm{TX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x}---\left(3\right)$

Wherein, TX _ibe i-th training sample, N is number of training, for the average of training sample, X is the training sample after standardization.σ _xrepresent the standard deviation of training sample, σ ² _xrepresent the variance of training sample.

Fuzzy neural network module, to passing the input variable of coming from data preprocessing module, carries out fuzzy reasoning and sets up fuzzy rule.Carry out fuzzy classification to what come from data preprocessing module biography through pretreated training sample X, obtain center and the width of each fuzzy clustering in fuzzy rule base.If the training sample X after p standardization _p=[X _p1..., X _pn], wherein n is the number of input variable.

If fuzzy neural network has R fuzzy rule, in order to try to achieve each fuzzy rule for training sample X _peach input variable X _pj, j=1 ..., n, obfuscation equation below will obtain its degree of membership to i-th fuzzy rule:

$M_{\mathrm{ij}}=\exp \{-\frac{{(X_{\mathrm{pj}}-m_{\mathrm{ij}})}^2}{{\mathrm{\σ}}_{\mathrm{ij}}^2}\}---\left(4\right)$

Wherein m _ijand σ _ijrepresent center and the width of jth Gauss's member function of i-th fuzzy rule respectively, tried to achieve by fuzzy clustering.

If the training sample X after standardization _pbe μ to the fitness of fuzzy rule i ⁽ⁱ⁾(X _p), then μ ⁽ⁱ⁾(X _p) large I determined by following formula:

${\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)=\underset{j=1}{\overset{n}{\mathrm{\Π}}}{M}_{\mathrm{ij}}\left(X_p\right)=\exp \{-\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\frac{{(X_{\mathrm{pj}}-m_{\mathrm{ij}})}^2}{{\mathrm{\σ}}_{\mathrm{ij}}^2}\}---\left(5\right)$

After trying to achieve the fitness of input training sample for each rule, fuzzy neural network exports fuzzy rule and derives to obtain last analytic solution.In conventional structure of fuzzy neural network, the process that each fuzzy rule is derived can be expressed as: the linear sum of products of first trying to achieve all input variables in training sample, then with the relevance grade μ of this linear sum of products with rule ⁽ⁱ⁾(X _p) be multiplied, obtain the output of final every bar fuzzy rule.The derivation of fuzzy rule i exports and can be expressed as follows:

$f^{\left(i\right)}={\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})---\left(6\right)$

${\hat{y}}_p=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}{f}^{\left(i\right)}+b=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}[{\mathrm{\μ}}^i\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})]+b---\left(7\right)$

In formula, f ⁽ⁱ⁾be the output of i-th fuzzy rule, that fuzzy neural network model exports the prediction of p training sample, a _ij, j=1 ..., n is the linear coefficient of a jth variable in i-th fuzzy rule, a _i0be the constant term of the linear sum of products of input variable in i-th fuzzy rule, b is output offset amount.

Support vector machine optimizes module, in formula (7), the determination of the parameter in the linear sum of products of input variable is the subject matter used during fuzzy neural network uses, here we adopt and original fuzzy rule derivation output form are converted to support vector machine optimization problem, re-use support vector machine and carry out linear optimization, transfer process is as follows:

$\begin{array}{c}{\hat{y}}_p=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}{f}^{\left(i\right)}+b\\ =\underset{i=1}{\overset{R}{\mathrm{\Σ}}}[{\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})]+b\\ =\underset{i=1}{\overset{R}{\mathrm{\Σ}}}\underset{j=0}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×{\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×X_{\mathrm{pj}}+b\end{array}---\left(8\right)$

Wherein X _p01 is constantly equal to for constant term.Order

Wherein, represent the reformulations of former training sample, namely original training sample be converted to as above formula form, the training sample as support vector machine:

Wherein y ₁..., y _nbe the target output of training sample, get S as new input training sample set, so original problem can be converted into following support vector machine primal-dual optimization problem:

$R(\mathrm{\ω},b)=\mathrm{\γ}\frac{1}{N}{\mathrm{\Σ}}_{p=1}^N{L}_{\mathrm{\ϵ}}(y_p,f\left(X_p\right))+\frac{1}{2}{\mathrm{\ω}}^T\mathrm{\ω}---\left(11\right)$

Wherein y _pinput training sample X _ptarget export, ω is the normal vector of support vector machine lineoid, f (X _p) be correspond to X _pmodel export, γ is the penalty factor of support vector machine, and R (ω, b) is the objective function of optimization problem, and N is number of training, L _ε(y _p, f (X _p)) expression formula is as follows:

Wherein ε is the error margin of optimization problem, and the forecast of the optimum derivation linear dimensions of fuzzy rule and primal-dual optimization problem that next use support vector machine to try to achieve fuzzy neural network exports:

$a_{\mathrm{ij}}=\underset{k=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\α}}_k^{*}-{\mathrm{\α}}_k){\mathrm{\μ}}^{\left(i\right)}{X}_{\mathrm{kj}}=\underset{k\∈\mathrm{SV}}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\α}}_k^{*}-{\mathrm{\α}}_k){\mathrm{\μ}}^{\left(i\right)}{X}_{\mathrm{kj}},i=1,...,R;j=0,...,n---\left(13\right)$

Wherein α _k, y respectively _p-f (X _p) be greater than 0 and be less than 0 time corresponding Lagrange multiplier, be the training sample X after corresponding to p standardization _pmI predicted value.

As preferred a kind of scheme, the industrial melt index soft measurement model of described optimum FUZZY NETWORK also comprises: model modification module, for the online updating of model, is regularly input in training set by off-line analysis data, upgrades fuzzy neural network model.

An industrial melt index soft measuring method for optimum FUZZY NETWORK, described flexible measurement method specific implementation step is as follows:

1), to propylene polymerization production process object, according to industrial analysis and Operations Analyst, select performance variable and easily survey the input of variable as model, performance variable and easily survey variable are obtained by DCS database;

2), for the model training sample will inputted from DCS database carry out pre-service, make the average of training sample be 0, variance is 1, and this process adopts following formula process:

Computation of mean values: $\stackrel{\‾}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{TX}}_i---\left(1\right)$

Calculate variance: ${\mathrm{\σ}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{TX}}_i-\stackrel{\‾}{\mathrm{TX}})---\left(2\right)$

Standardization: $X=\frac{\mathrm{TX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x}---\left(3\right)$

Wherein, TX _ibe i-th training sample, N is number of training, for the average of training sample, X is the training sample after standardization.σ _xrepresent the standard deviation of training sample, σ ² _xrepresent the variance of training sample.

3), to passing the input variable of coming from data preprocessing module, carry out fuzzy reasoning and set up fuzzy rule.Carry out fuzzy classification to what come from data preprocessing module biography through pretreated training sample X, obtain center and the width of each fuzzy clustering in fuzzy rule base.If the training sample X after p standardization _p=[X _p1..., X _pn], wherein n is the number of input variable.

If fuzzy neural network has R fuzzy rule, in order to try to achieve each fuzzy rule for training sample X _peach input variable X _pj, j=1 ..., n, obfuscation equation below will obtain its degree of membership to i-th fuzzy rule:

$M_{\mathrm{ij}}=\exp \{-\frac{{(X_{\mathrm{pj}}-m_{\mathrm{ij}})}^2}{{\mathrm{\σ}}_{\mathrm{ij}}^2}\}---\left(4\right)$

Wherein m _ijand σ _ijrepresent center and the width of jth Gauss's member function of i-th fuzzy rule respectively, tried to achieve by fuzzy clustering.

If the training sample X after standardization _pbe μ to the fitness of fuzzy rule i ⁽ⁱ⁾(X _p), then μ ⁽ⁱ⁾(X _p) large I determined by following formula:

${\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)=\underset{j=1}{\overset{n}{\mathrm{\Π}}}{M}_{\mathrm{ij}}\left(X_p\right)=\exp \{-\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\frac{{(X_{\mathrm{pj}}-m_{\mathrm{ij}})}^2}{{\mathrm{\σ}}_{\mathrm{ij}}^2}\}---\left(5\right)$

After trying to achieve the fitness of input training sample for each rule, fuzzy neural network exports fuzzy rule and derives to obtain last analytic solution.In conventional structure of fuzzy neural network, the process that each fuzzy rule is derived can be expressed as: the linear sum of products of first trying to achieve all input variables in training sample, then with the relevance grade μ of this linear sum of products with rule ⁱ(X _p) be multiplied, obtain the output of final every bar fuzzy rule.The derivation of fuzzy rule i exports and can be expressed as follows:

$f^{\left(i\right)}={\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})---\left(6\right)$

${\hat{y}}_p=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}{f}^{\left(i\right)}+b=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}[{\mathrm{\μ}}^i\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})]+b---\left(7\right)$

In formula, f ⁽ⁱ⁾be the output of i-th fuzzy rule, that fuzzy neural network model exports the prediction of p training sample, a _ij, j=1 ..., n is the linear coefficient of a jth variable in i-th fuzzy rule, a _i0be the constant term of the linear sum of products of input variable in i-th fuzzy rule, b is output offset amount.

4), in formula (7), the determination of the parameter in the linear sum of products of input variable is the subject matter used during fuzzy neural network uses, here we adopt and original fuzzy rule derivation output form are converted to support vector machine optimization problem, re-use support vector machine and carry out linear optimization, transfer process is as follows:

$\begin{array}{c}{\hat{y}}_p=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}{f}^{\left(i\right)}+b\\ =\underset{i=1}{\overset{R}{\mathrm{\Σ}}}[{\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})]+b\\ =\underset{i=1}{\overset{R}{\mathrm{\Σ}}}\underset{j=0}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×{\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×X_{\mathrm{pj}}+b\end{array}---\left(8\right)$

Wherein X _p01 is constantly equal to for constant term.Order

Wherein, represent the reformulations of former training sample, namely original training sample be converted to as above formula form, the training sample as support vector machine:

Wherein y ₁..., y _nbe the target output of training sample, get S as new input training sample set, so original problem can be converted into following support vector machine primal-dual optimization problem:

$R(\mathrm{\ω},b)=\mathrm{\γ}\frac{1}{N}{\mathrm{\Σ}}_{p=1}^N{L}_{\mathrm{\ϵ}}(y_p,f\left(X_p\right))+\frac{1}{2}{\mathrm{\ω}}^T\mathrm{\ω}---\left(11\right)$

Wherein y _pinput training sample X _ptarget export, ω is the normal vector of support vector machine lineoid, f (X _p) be correspond to X _pmodel export, γ is the penalty factor of support vector machine, and R (ω, b) is the objective function of optimization problem, and N is number of training, L _ε(y _p, f (X _p)) expression formula is as follows:

Wherein ε is the error margin of optimization problem, and the forecast of the optimum derivation linear dimensions of fuzzy rule and primal-dual optimization problem that next use support vector machine to try to achieve fuzzy neural network exports:

$a_{\mathrm{ij}}=\underset{k=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\α}}_k^{*}-{\mathrm{\α}}_k){\mathrm{\μ}}^{\left(i\right)}{X}_{\mathrm{kj}}=\underset{k\∈\mathrm{SV}}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\α}}_k^{*}-{\mathrm{\α}}_k){\mathrm{\μ}}^{\left(i\right)}{X}_{\mathrm{kj}},i=1,...,R;j=0,...,n---\left(13\right)$

Wherein α _k, y respectively _p-f (X _p) be greater than 0 and be less than 0 time corresponding Lagrange multiplier, be the training sample X after corresponding to p standardization _pmI predicted value.

As preferred a kind of scheme: described flexible measurement method is further comprising the steps of: 5), regularly off-line analysis data is input to training sample and concentrates, upgrade fuzzy neural network model.

Technical conceive of the present invention is: carry out online soft sensor to the important quality index melting index of propylene polymerization production process, overcome the deficiency that existing polypropylene melting index measurement instrument measuring accuracy is not high, appearance makes an uproar ability, model parameter setting difficulty is large, introduce support vector machine and Automatic Optimal is carried out to fuzzy neural network model.This model has following advantage relative to existing melting index soft-sensing model: (1) reduces noise and manual operation error to the impact of model prediction precision; (2) enhance the popularization performance of model, the Expired Drugs of existing model is effectively suppressed; (3) improve the stability of model, reduce model and cross the possibility that over-fitting occurs.

Beneficial effect of the present invention is mainly manifested in: 1, on-line measurement; 2, model upgrades automatically; 3, anti-noise jamming ability strong, 4, precision is high; 5, Generalization Ability is strong.

Accompanying drawing explanation

Fig. 1 is the industrial melt index soft measurement instrument of optimum FUZZY NETWORK and the basic structure schematic diagram of method;

Fig. 2 is the industrial melt index soft measurement model structural representation of optimum FUZZY NETWORK.

Embodiment

Below in conjunction with accompanying drawing, the invention will be further described.The embodiment of the present invention is used for explaining and the present invention is described, instead of limits the invention, and in the protection domain of spirit of the present invention and claim, any amendment make the present invention and change, all fall into protection scope of the present invention.

Embodiment 1

With reference to Fig. 1, Fig. 2, a kind of propylene polymerization production process soft measuring instrument based on support vector machine Optimization of Fuzzy neural network, comprise propylene polymerization production process 1, for measuring the field intelligent instrument 2 easily surveying variable, for measuring the control station 3 of performance variable, the DCS database 4 of store data and melt index flexible measured value display instrument 6, described field intelligent instrument 2, control station 3 is connected with propylene polymerization production process 1, described field intelligent instrument 2, control station 3 is connected with DCS database 4, described soft measuring instrument also comprises the soft-sensing model 5 of support vector machine Optimization of Fuzzy neural network, described DCS database 4 is connected with the input end of the industrial melt index soft measurement model 5 of described optimum FUZZY NETWORK, the output terminal of the industrial melt index soft measurement model 5 of described optimum FUZZY NETWORK is connected with melt index flexible measured value display instrument 6, the industrial melt index soft measurement model of described optimum FUZZY NETWORK comprises:

Data preprocessing module: the model training sample for inputting from DCS database carries out pre-service, make the average of training sample be 0, variance is 1, and this process adopts following formula process:

Computation of mean values: $\stackrel{\‾}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{TX}}_i---\left(1\right)$

Calculate variance: ${\mathrm{\σ}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{TX}}_i-\stackrel{\‾}{\mathrm{TX}})---\left(2\right)$

Standardization: $X=\frac{\mathrm{TX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x}---\left(3\right)$

Wherein, TX _ibe i-th training sample, N is number of training, for the average of training sample, X is the training sample after standardization.σ _xrepresent the standard deviation of training sample, σ ² _xrepresent the variance of training sample.

Fuzzy neural network module, to passing the input variable of coming from data preprocessing module, carries out fuzzy reasoning and sets up fuzzy rule.Carry out fuzzy classification to what come from data preprocessing module biography through pretreated training sample X, obtain center and the width of each fuzzy clustering in fuzzy rule base.If the training sample X after p standardization _p=[X _p1..., X _pn], wherein n is the number of input variable.

If fuzzy neural network has R fuzzy rule, in order to try to achieve each fuzzy rule for training sample X _peach input variable X _pj, j=1 ..., n, obfuscation equation below will obtain its degree of membership to i-th fuzzy rule:

$M_{\mathrm{ij}}=\exp \{-\frac{{(X_{\mathrm{pj}}-m_{\mathrm{ij}})}^2}{{\mathrm{\σ}}_{\mathrm{ij}}^2}\}---\left(4\right)$

Wherein m _ijand σ _ijrepresent center and the width of jth Gauss's member function of i-th fuzzy rule respectively, tried to achieve by fuzzy clustering.

If the training sample X after standardization _pbe μ to the fitness of fuzzy rule i ⁽ⁱ⁾(X _p), then μ ⁽ⁱ⁾(X _p) large I determined by following formula:

${\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)=\underset{j=1}{\overset{n}{\mathrm{\Π}}}{M}_{\mathrm{ij}}\left(X_p\right)=\exp \{-\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\frac{{(X_{\mathrm{pj}}-m_{\mathrm{ij}})}^2}{{\mathrm{\σ}}_{\mathrm{ij}}^2}\}---\left(5\right)$

After trying to achieve the fitness of input training sample for each rule, fuzzy neural network exports fuzzy rule and derives to obtain last analytic solution.In conventional structure of fuzzy neural network, the process that each fuzzy rule is derived can be expressed as: the linear sum of products of first trying to achieve all input variables in training sample, then with the relevance grade μ of this linear sum of products with rule ⁽ⁱ⁾(X _p) be multiplied, obtain the output of final every bar fuzzy rule.The derivation of fuzzy rule i exports and can be expressed as follows:

$f^{\left(i\right)}={\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})---\left(6\right)$

${\hat{y}}_p=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}{f}^{\left(i\right)}+b=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}[{\mathrm{\μ}}^i\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})]+b---\left(7\right)$

In formula, f ⁽ⁱ⁾be the output of i-th fuzzy rule, that fuzzy neural network model exports the prediction of p training sample, a _ij, j=1 ..., n is the linear coefficient of a jth variable in i-th fuzzy rule, a _i0be the constant term of the linear sum of products of input variable in i-th fuzzy rule, b is output offset amount.

Support vector machine optimizes module, in formula (7), the determination of the parameter in the linear sum of products of input variable is the subject matter used during fuzzy neural network uses, here we adopt and original fuzzy rule derivation output form are converted to support vector machine optimization problem, re-use support vector machine and carry out linear optimization, transfer process is as follows:

$\begin{array}{c}{\hat{y}}_p=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}{f}^{\left(i\right)}+b\\ =\underset{i=1}{\overset{R}{\mathrm{\Σ}}}[{\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})]+b\\ =\underset{i=1}{\overset{R}{\mathrm{\Σ}}}\underset{j=0}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×{\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×X_{\mathrm{pj}}+b\end{array}---\left(8\right)$

Wherein X _p01 is constantly equal to for constant term.Order

Wherein, represent the reformulations of former training sample, namely original training sample be converted to as above formula form, the training sample as support vector machine:

Wherein y ₁..., y _nbe the target output of training sample, get S as new input training sample set, so original problem can be converted into following support vector machine primal-dual optimization problem:

$R(\mathrm{\ω},b)=\mathrm{\γ}\frac{1}{N}{\mathrm{\Σ}}_{p=1}^N{L}_{\mathrm{\ϵ}}(y_p,f\left(X_p\right))+\frac{1}{2}{\mathrm{\ω}}^T\mathrm{\ω}---\left(11\right)$

Wherein y _pinput training sample X _ptarget export, ω is the normal vector of support vector machine lineoid, f (X _p) be correspond to X _pmodel export, γ is the penalty factor of support vector machine, and R (ω, b) is the objective function of optimization problem, and N is number of training, L _ε(y _p, f (X _p)) expression formula is as follows:

Wherein ε is the error margin of optimization problem, and the forecast of the optimum derivation linear dimensions of fuzzy rule and primal-dual optimization problem that next use support vector machine to try to achieve fuzzy neural network exports:

$a_{\mathrm{ij}}=\underset{k=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\α}}_k^{*}-{\mathrm{\α}}_k){\mathrm{\μ}}^{\left(i\right)}{X}_{\mathrm{kj}}=\underset{k\∈\mathrm{SV}}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\α}}_k^{*}-{\mathrm{\α}}_k){\mathrm{\μ}}^{\left(i\right)}{X}_{\mathrm{kj}},i=1,...,R;j=0,...,n---\left(13\right)$

Wherein α _k, y respectively _p-f (X _p) be greater than 0 and be less than 0 time corresponding Lagrange multiplier, be the training sample X after corresponding to p standardization _pmI predicted value.

The industrial melt index soft measurement model of described optimum FUZZY NETWORK also comprises: model modification module, for the online updating of model, is regularly input in training set by off-line analysis data, upgrades fuzzy neural network model.

According to reaction mechanism and flow process analysis, consider in polypropylene production process the various factors that melting index has an impact, get nine performance variables conventional in actual production process and easily survey variable as modeling variable, have: three strand of third rare feed flow rates, major catalyst flow rate, cocatalyst flow rate, temperature in the kettle, pressure, liquid level, hydrogen volume concentration in still.Table 1 lists 9 the modeling variablees inputted as soft-sensing model 5, to be respectively in temperature in the kettle (T), still in pressure (P), still hydrogen volume concentration (X in liquid level (L), still _v), 3 bursts of propylene feed flow rates (first strand of third rare feed flow rates f1, second strand of third rare feed flow rates f2, the 3rd strand of third rare feed flow rates f3), 2 bursts of catalyst charge flow rates (major catalyst flow rate f4, cocatalyst flow rate f5).Polyreaction in reactor is that reaction mass mixes rear participation reaction repeatedly, and therefore mode input variable relates to the mean value in process variable employing front some moment of material.The data acquisition mean value of last hour in this example.Melting index off-line laboratory values is as the output variable of soft-sensing model 5.Obtained by hand sampling, off-line assay, within every 4 hours, analyze and gather once.

Field intelligent instrument 2 and control station 3 are connected with propylene polymerization production process 1, are connected with DCS database 4; Soft-sensing model 5 is connected with DCS database and hard measurement value display instrument 6.Field intelligent instrument 2 measures the easy survey variable that propylene polymerization produces object, is transferred to DCS database 4 by easily surveying variable; Control station 3 controls the performance variable that propylene polymerization produces object, performance variable is transferred to DCS database 4.The variable data recorded in DCS database 4 is as the input of the soft-sensing model 5 based on particle cluster algorithm optimization Weighted Least Squares Support Vector Machines fuzzifying equation, hard measurement value display instrument 6 for showing the output of the industrial melt index soft measurement model 5 of optimum FUZZY NETWORK, i.e. hard measurement value.

Table 1: modeling variable needed for the industrial melt index soft measurement model of optimum FUZZY NETWORK

Variable symbol	Variable implication	Variable symbol	Variable implication
				T	Temperature in the kettle	f1	First strand of third rare feed flow rates
P	Pressure in still	f2	Second strand of third rare feed flow rates
				L	Liquid level in still	f3	3rd strand of third rare feed flow rates
X v	Hydrogen volume concentration in still	f4	Major catalyst flow rate
						f5	Cocatalyst flow rate

The industrial melt index soft measurement model 5 of optimum FUZZY NETWORK, comprises following 3 parts:

Data preprocessing module 7 carries out pre-service for the model training sample will inputted from DCS database, and make the average of training sample be 0, variance is 1, and this process adopts following formula process:

Computation of mean values: $\stackrel{\‾}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{TX}}_i---\left(1\right)$

Calculate variance: ${\mathrm{\σ}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{TX}}_i-\stackrel{\‾}{\mathrm{TX}})---\left(2\right)$

Standardization: $X=\frac{\mathrm{TX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x}---\left(3\right)$

Wherein, TX _ibe i-th training sample, N is number of training, for the average of training sample, X is the training sample after standardization.σ _xrepresent the standard deviation of training sample, σ ² _xrepresent the variance of training sample.

Fuzzy neural network module 8, to passing the input variable of coming from data preprocessing module, carries out fuzzy reasoning and sets up fuzzy rule.Carry out fuzzy classification to what come from data preprocessing module biography through pretreated training sample X, obtain center and the width of each fuzzy clustering in fuzzy rule base.If the training sample X after p standardization _p=[X _p1..., X _pn], wherein n is the number of input variable.

If fuzzy neural network has R fuzzy rule, in order to try to achieve each fuzzy rule for training sample X _peach input variable X _pj, j=1 ..., n, obfuscation equation below will obtain its degree of membership to i-th fuzzy rule:

$M_{\mathrm{ij}}=\exp \{-\frac{{(X_{\mathrm{pj}}-m_{\mathrm{ij}})}^2}{{\mathrm{\σ}}_{\mathrm{ij}}^2}\}---\left(4\right)$

Wherein m _ijand σ _ijrepresent center and the width of jth Gauss's member function of i-th fuzzy rule respectively, tried to achieve by fuzzy clustering.

If the training sample X after standardization _pbe μ to the fitness of fuzzy rule i ⁽ⁱ⁾(X _p), then μ ⁽ⁱ⁾(X _p) large I determined by following formula:

${\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)=\underset{j=1}{\overset{n}{\mathrm{\Π}}}{M}_{\mathrm{ij}}\left(X_p\right)=\exp \{-\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\frac{{(X_{\mathrm{pj}}-m_{\mathrm{ij}})}^2}{{\mathrm{\σ}}_{\mathrm{ij}}^2}\}---\left(5\right)$

After trying to achieve the fitness of input training sample for each rule, fuzzy neural network exports fuzzy rule and derives to obtain last analytic solution.In conventional structure of fuzzy neural network, the process that each fuzzy rule is derived can be expressed as: the linear sum of products of first trying to achieve all input variables in training sample, then with the relevance grade μ of this linear sum of products with rule ⁽ⁱ⁾(X _p) be multiplied, obtain the output of final every bar fuzzy rule.The derivation of fuzzy rule i exports and can be expressed as follows:

$f^{\left(i\right)}={\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})---\left(6\right)$

${\hat{y}}_p=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}{f}^{\left(i\right)}+b=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}[{\mathrm{\μ}}^i\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})]+b---\left(7\right)$

In formula, f ⁽ⁱ⁾be the output of i-th fuzzy rule, that fuzzy neural network model exports the prediction of p training sample, a _ij, j=1 ..., n is the linear coefficient of a jth variable in i-th fuzzy rule, a _i0be the constant term of the linear sum of products of input variable in i-th fuzzy rule, b is output offset amount.

Support vector machine optimizes module 9, in formula (7), the determination of the parameter in the linear sum of products of input variable is the subject matter used during fuzzy neural network uses, here we adopt and original fuzzy rule derivation output form are converted to support vector machine optimization problem, re-use support vector machine and carry out linear optimization, transfer process is as follows:

$\begin{array}{c}{\hat{y}}_p=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}{f}^{\left(i\right)}+b\\ =\underset{i=1}{\overset{R}{\mathrm{\Σ}}}[{\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})]+b\\ =\underset{i=1}{\overset{R}{\mathrm{\Σ}}}\underset{j=0}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×{\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×X_{\mathrm{pj}}+b\end{array}---\left(8\right)$

Wherein X _p01 is constantly equal to for constant term.Order

Wherein, represent the reformulations of former training sample, namely original training sample be converted to as above formula form, the training sample as support vector machine:

Wherein y ₁..., y _nbe the target output of training sample, get S as new input training sample set, so original problem can be converted into following support vector machine primal-dual optimization problem:

$R(\mathrm{\ω},b)=\mathrm{\γ}\frac{1}{N}{\mathrm{\Σ}}_{p=1}^N{L}_{\mathrm{\ϵ}}(y_p,f\left(X_p\right))+\frac{1}{2}{\mathrm{\ω}}^T\mathrm{\ω}---\left(11\right)$

Wherein y _pinput training sample X _ptarget export, ω is the normal vector of support vector machine lineoid, f (X _p) be correspond to X _pmodel export, γ is the penalty factor of support vector machine, and R (ω, b) is the objective function of optimization problem, and N is number of training, L _ε(y _p, f (X _p)) expression formula is as follows:

Wherein ε is the error margin of optimization problem, and the forecast of the optimum derivation linear dimensions of fuzzy rule and primal-dual optimization problem that next use support vector machine to try to achieve fuzzy neural network exports:

$a_{\mathrm{ij}}=\underset{k=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\α}}_k^{*}-{\mathrm{\α}}_k){\mathrm{\μ}}^{\left(i\right)}{X}_{\mathrm{kj}}=\underset{k\∈\mathrm{SV}}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\α}}_k^{*}-{\mathrm{\α}}_k){\mathrm{\μ}}^{\left(i\right)}{X}_{\mathrm{kj}},i=1,...,R;j=0,...,n---\left(13\right)$

Wherein α _k, y respectively _p-f (X _p) be greater than 0 and be less than 0 time corresponding Lagrange multiplier, be the training sample X after corresponding to p standardization _pmI predicted value.

Model modification module 10, for the online updating of model, is regularly input in training set by off-line analysis data, upgrades fuzzy neural network model.

Embodiment 2

With reference to Fig. 1, Fig. 2, a kind of industrial melt index soft measuring method of optimum FUZZY NETWORK, described flexible measurement method specific implementation step is as follows:

1), to propylene polymerization production process object, according to industrial analysis and Operations Analyst, select performance variable and easily survey the input of variable as model, performance variable and easily survey variable are obtained by DCS database;

2), for the model training sample will inputted from DCS database carry out pre-service, make the average of training sample be 0, variance is 1, and this process adopts following formula process:

Computation of mean values: $\stackrel{\‾}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{TX}}_i---\left(1\right)$

Calculate variance: ${\mathrm{\σ}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{TX}}_i-\stackrel{\‾}{\mathrm{TX}})---\left(2\right)$

Standardization: $X=\frac{\mathrm{TX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x}---\left(3\right)$

Wherein, TX _ibe i-th training sample, N is number of training, for the average of training sample, X is the training sample after standardization.σ _xrepresent the standard deviation of training sample, σ ² _xrepresent the variance of training sample.

3), to passing the input variable of coming from data preprocessing module, carry out fuzzy reasoning and set up fuzzy rule.Carry out fuzzy classification to what come from data preprocessing module biography through pretreated training sample X, obtain center and the width of each fuzzy clustering in fuzzy rule base.If the training sample X after p standardization _p=[X _p1..., X _pn], wherein n is the number of input variable.

If fuzzy neural network has R fuzzy rule, in order to try to achieve each fuzzy rule for training sample X _peach input variable X _pj, j=1 ..., n, obfuscation equation below will obtain its degree of membership to i-th fuzzy rule:

$M_{\mathrm{ij}}=\exp \{-\frac{{(X_{\mathrm{pj}}-m_{\mathrm{ij}})}^2}{{\mathrm{\σ}}_{\mathrm{ij}}^2}\}---\left(4\right)$

Wherein m _ijand σ _ijrepresent center and the width of jth Gauss's member function of i-th fuzzy rule respectively, tried to achieve by fuzzy clustering.

If the training sample X after standardization _pbe μ to the fitness of fuzzy rule i ⁽ⁱ⁾(X _p), then μ ⁽ⁱ⁾(X _p) large I determined by following formula:

${\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)=\underset{j=1}{\overset{n}{\mathrm{\Π}}}{M}_{\mathrm{ij}}\left(X_p\right)=\exp \{-\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\frac{{(X_{\mathrm{pj}}-m_{\mathrm{ij}})}^2}{{\mathrm{\σ}}_{\mathrm{ij}}^2}\}---\left(5\right)$

After trying to achieve the fitness of input training sample for each rule, fuzzy neural network exports fuzzy rule and derives to obtain last analytic solution.In conventional structure of fuzzy neural network, the process that each fuzzy rule is derived can be expressed as: the linear sum of products of first trying to achieve all input variables in training sample, then with the relevance grade μ of this linear sum of products with rule ⁽ⁱ⁾(X _p) be multiplied, obtain the output of final every bar fuzzy rule.The derivation of fuzzy rule i exports and can be expressed as follows:

$f^{\left(i\right)}={\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})---\left(6\right)$

${\hat{y}}_p=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}{f}^{\left(i\right)}+b=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}[{\mathrm{\μ}}^i\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})]+b---\left(7\right)$

In formula, f ⁽ⁱ⁾be the output of i-th fuzzy rule, that fuzzy neural network model exports the prediction of p training sample, a _ij, j=1 ..., n is the linear coefficient of a jth variable in i-th fuzzy rule, a _i0be the constant term of the linear sum of products of input variable in i-th fuzzy rule, b is output offset amount.

4), in formula (7), the determination of the parameter in the linear sum of products of input variable is the subject matter used during fuzzy neural network uses, here we adopt and original fuzzy rule derivation output form are converted to support vector machine optimization problem, re-use support vector machine and carry out linear optimization, transfer process is as follows:

$\begin{array}{c}{\hat{y}}_p=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}{f}^{\left(i\right)}+b\\ =\underset{i=1}{\overset{R}{\mathrm{\Σ}}}[{\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})]+b\\ =\underset{i=1}{\overset{R}{\mathrm{\Σ}}}\underset{j=0}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×{\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×X_{\mathrm{pj}}+b\end{array}---\left(8\right)$

Wherein X _p01 is constantly equal to for constant term.Order

Wherein, represent the reformulations of former training sample, namely original training sample be converted to as above formula form, the training sample as support vector machine:

Wherein y ₁..., y _nbe the target output of training sample, get S as new input training sample set, so original problem can be converted into following support vector machine primal-dual optimization problem:

$R(\mathrm{\ω},b)=\mathrm{\γ}\frac{1}{N}{\mathrm{\Σ}}_{p=1}^N{L}_{\mathrm{\ϵ}}(y_p,f\left(X_p\right))+\frac{1}{2}{\mathrm{\ω}}^T\mathrm{\ω}---\left(11\right)$

Wherein y _pinput training sample X _ptarget export, ω is the normal vector of support vector machine lineoid, f (X _p) be correspond to X _pmodel export, γ is the penalty factor of support vector machine, and R (ω, b) is the objective function of optimization problem, and N is number of training, L _ε(y _p, f (X _p)) expression formula is as follows:

Wherein ε is the error margin of optimization problem, and the forecast of the optimum derivation linear dimensions of fuzzy rule and primal-dual optimization problem that next use support vector machine to try to achieve fuzzy neural network exports:

$a_{\mathrm{ij}}=\underset{k=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\α}}_k^{*}-{\mathrm{\α}}_k){\mathrm{\μ}}^{\left(i\right)}{X}_{\mathrm{kj}}=\underset{k\∈\mathrm{SV}}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\α}}_k^{*}-{\mathrm{\α}}_k){\mathrm{\μ}}^{\left(i\right)}{X}_{\mathrm{kj}},i=1,...,R;j=0,...,n---\left(13\right)$

Wherein α _k, y respectively _p-f (X _p) be greater than 0 and be less than 0 time corresponding Lagrange multiplier, be the training sample X after corresponding to p standardization _pmI predicted value.

As preferred a kind of scheme: described flexible measurement method is further comprising the steps of: 5), regularly off-line analysis data is input to training sample and concentrates, upgrade fuzzifying equation model.

The method specific implementation step of the present embodiment is as follows:

Step 1: to propylene polymerization production process object 1, according to industrial analysis and Operations Analyst, selects performance variable and easily surveys the input of variable as model.Performance variable and easily survey variable are obtained by DCS database 4.

Step 2: carry out pre-service to training sample, is completed by data preprocessing module 7.

Step 3: set up initial fuzzy neural network model 8 based on pretreated training sample data.Input data obtain as described in step 2, export data and chemically examine acquisition by off-line.

Step 4: optimize the Anti-fuzzy output parameter that module 9 optimizes initial fuzzy neural network model 8 by support vector machine.

Step 5: off-line analysis data is regularly input in training set by model modification module 10, upgrade fuzzy neural network model, the industrial melt index soft measurement model 5 of optimum FUZZY NETWORK has been set up.

Step 6: melt index flexible measured value display instrument 6 shows the output of the industrial melt index soft measurement model 5 of optimum FUZZY NETWORK, completes the display of measuring industrial polypropylene producing melt index flexible.

Claims (2)

1. the industrial melt index soft measurement instrument of an optimum FUZZY NETWORK, comprise for measuring the field intelligent instrument easily surveying variable, for measuring the control station of performance variable, the DCS database of store data and melt index flexible measured value display instrument, described field intelligent instrument, control station is connected with DCS database, it is characterized in that: described soft measuring instrument also comprises the industrial melt index soft measurement model of optimum FUZZY NETWORK, described DCS database is connected with the input end of the industrial melt index soft measurement model of described optimum FUZZY NETWORK, the output terminal of the industrial melt index soft measurement model of described optimum FUZZY NETWORK is connected with melt index flexible measured value display instrument, the industrial melt index soft measurement model of described optimum FUZZY NETWORK comprises:

Data preprocessing module, carries out pre-service for the model training sample will inputted from DCS database, and make the average of training sample be 0, variance is 1, and this process adopts following formula process:

Computation of mean values: $\stackrel{\‾}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{TX}}_i---\left(1\right)$

Calculate variance: ${\mathrm{\σ}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{TX}}_i-\stackrel{\‾}{\mathrm{TX}})---\left(2\right)$

Standardization: $X=\frac{\mathrm{TX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x}---\left(3\right)$

Wherein, TX _ibe i-th training sample, N is number of training, for the average of training sample, X is the training sample after standardization; σ _xrepresent the standard deviation of training sample, σ ² _xrepresent the variance of training sample;

Fuzzy neural network module, to passing the input variable of coming from data preprocessing module, carries out fuzzy reasoning and sets up fuzzy rule; Carry out fuzzy classification to what come from data preprocessing module biography through pretreated training sample X, obtain center and the width of each fuzzy clustering in fuzzy rule base; If the training sample X after p standardization _p=[X _p1..., X _pn], wherein n is the number of input variable;

If fuzzy neural network has R fuzzy rule, in order to try to achieve each fuzzy rule for training sample X _peach input variable X _pj, j=1 ..., n, obfuscation equation below will obtain its degree of membership to i-th fuzzy rule:

$M_{\mathrm{ij}}=\exp \{-\frac{{(X_{\mathrm{pj}}-m_{\mathrm{ij}})}^2}{{\mathrm{\σ}}_{\mathrm{ij}}^2}\}---\left(4\right)$

Wherein m _ijand σ _ijrepresent center and the width of jth Gauss's member function of i-th fuzzy rule respectively, tried to achieve by fuzzy clustering;

If the training sample X after standardization _pbe μ to the fitness of fuzzy rule i ⁽ⁱ⁾(X _p), then μ ⁽ⁱ⁾(X _p) large I determined by following formula:

${\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)=\underset{j=1}{\overset{n}{\mathrm{\Π}}}{M}_{\mathrm{ij}}\left(X_p\right)=\exp \{-\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\frac{{(X_{\mathrm{pj}}-m_{\mathrm{ij}})}^2}{{\mathrm{\σ}}_{\mathrm{ij}}^2}\}---\left(5\right)$

After trying to achieve the fitness of input training sample for each rule, fuzzy neural network exports fuzzy rule and derives to obtain last analytic solution; In conventional structure of fuzzy neural network, the process that each fuzzy rule is derived can be expressed as: the linear sum of products of first trying to achieve all input variables in training sample, then with the relevance grade μ of this linear sum of products with rule ⁱ(X _p) be multiplied, obtain the output of final every bar fuzzy rule; The derivation of fuzzy rule i exports and can be expressed as follows:

$f^{\left(i\right)}={\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})---\left(6\right)$

${\hat{y}}_p=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}{f}^{\left(i\right)}+b=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}[{\mathrm{\μ}}^i\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})]+b---\left(7\right)$

In formula, f ⁽ⁱ⁾be the output of i-th fuzzy rule, that fuzzy neural network model exports the prediction of p training sample, a _ij, j=1 ..., n is the linear coefficient of a jth variable in i-th fuzzy rule, a _i0be the constant term of the linear sum of products of input variable in i-th fuzzy rule, b is output offset amount;

Support vector machine optimizes module, in formula (7), the determination of the parameter in the linear sum of products of input variable is the subject matter used during fuzzy neural network uses, here we adopt and original fuzzy rule derivation output form are converted to support vector machine optimization problem, re-use support vector machine and carry out linear optimization, transfer process is as follows:

$\begin{array}{c}{\hat{y}}_p=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}{f}^{\left(i\right)}+b\\ =\underset{i=1}{\overset{R}{\mathrm{\Σ}}}[{\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})]+b\\ =\underset{i=1}{\overset{R}{\mathrm{\Σ}}}\underset{j=0}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×{\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×X_{\mathrm{pj}}+b\end{array}---\left(8\right)$

Wherein X _p01 is constantly equal to for constant term; Order

Wherein, represent the reformulations of former training sample, namely original training sample be converted to as above formula form, the training sample as support vector machine:

Wherein y ₁..., y _nbe the target output of training sample, get S as new input training sample set, so original problem can be converted into following support vector machine primal-dual optimization problem:

$R(\mathrm{\ω},b)=\mathrm{\γ}\frac{1}{N}{\mathrm{\Σ}}_{p=1}^N{L}_{\mathrm{\ϵ}}(y_p,f\left(X_p\right))+\frac{1}{2}{\mathrm{\ω}}^T\mathrm{\ω}---\left(11\right)$

Wherein y _pinput training sample X _ptarget export, ω is the normal vector of support vector machine lineoid, f (X _p) be correspond to X _pmodel export, γ is the penalty factor of support vector machine, and R (ω, b) is the objective function of optimization problem, and N is number of training, L _ε(y _p, f (X _p)) expression formula is as follows:

Wherein ε is the error margin of optimization problem, and the forecast of the optimum derivation linear dimensions of fuzzy rule and primal-dual optimization problem that next use support vector machine to try to achieve fuzzy neural network exports:

$a_{\mathrm{ij}}=\underset{k=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\α}}_k^{*}-{\mathrm{\α}}_k){\mathrm{\μ}}^{\left(i\right)}{X}_{\mathrm{kj}}=\underset{k\∈\mathrm{SV}}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\α}}_k^{*}-{\mathrm{\α}}_k){\mathrm{\μ}}^{\left(i\right)}{X}_{\mathrm{kj}},i=1,...,R;j=0,...,n---\left(13\right)$

Wherein α _k, (k=1 ..., N) and be y respectively _p-f (X _p) be greater than 0 and be less than 0 time corresponding Lagrange multiplier, be the training sample X after corresponding to p standardization _pmI predicted value;

The industrial melt index soft measurement model of described optimum FUZZY NETWORK also comprises:

Model modification module, for the online updating of model, is regularly input in training set by off-line analysis data, upgrades fuzzy neural network model.

2., with the flexible measurement method that the industrial melt index soft measurement instrument of a kind of optimum FUZZY NETWORK as claimed in claim 1 realizes, it is characterized in that: described flexible measurement method specific implementation step is as follows:

1), to propylene polymerization production process object, according to industrial analysis and Operations Analyst, select performance variable and easily survey the input of variable as model, performance variable and easily survey variable are obtained by DCS database;

2), for the model training sample will inputted from DCS database carry out pre-service, make the average of training sample be 0, variance is 1, and this process adopts following formula process:

Computation of mean values: $\stackrel{\‾}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{TX}}_i---\left(1\right)$

Calculate variance: ${\mathrm{\σ}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{TX}}_i-\stackrel{\‾}{\mathrm{TX}})---\left(2\right)$

Standardization: $X=\frac{\mathrm{TX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x}---\left(3\right)$

Wherein, TX _ibe i-th training sample, N is number of training, for the average of training sample, X is the training sample after standardization; σ _xrepresent the standard deviation of training sample, σ ² _xrepresent the variance of training sample;

3), to passing the input variable of coming from data preprocessing module, carry out fuzzy reasoning and set up fuzzy rule; Carry out fuzzy classification to what come from data preprocessing module biography through pretreated training sample X, obtain center and the width of each fuzzy clustering in fuzzy rule base; If the training sample X after p standardization _p=[X _p1..., X _pn], wherein n is the number of input variable;

If fuzzy neural network has R fuzzy rule, in order to try to achieve each fuzzy rule for training sample X _peach input variable X _pj, j=1 ..., n, obfuscation equation below will obtain its degree of membership to i-th fuzzy rule:

$M_{\mathrm{ij}}=\exp \{-\frac{{(X_{\mathrm{pj}}-m_{\mathrm{ij}})}^2}{{\mathrm{\σ}}_{\mathrm{ij}}^2}\}---\left(4\right)$

Wherein m _ijand σ _ijrepresent center and the width of jth Gauss's member function of i-th fuzzy rule respectively, tried to achieve by fuzzy clustering;

If the training sample X after standardization _pbe μ to the fitness of fuzzy rule i ⁽ⁱ⁾(X _p), then μ ⁽ⁱ⁾(X _p) large I determined by following formula:

${\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)=\underset{j=1}{\overset{n}{\mathrm{\Π}}}{M}_{\mathrm{ij}}\left(X_p\right)=\exp \{-\underset{j=1}{\overset{n}{\mathrm{\Σ}}}\frac{{(X_{\mathrm{pj}}-m_{\mathrm{ij}})}^2}{{\mathrm{\σ}}_{\mathrm{ij}}^2}\}---\left(5\right)$

After trying to achieve the fitness of input training sample for each rule, fuzzy neural network exports fuzzy rule and derives to obtain last analytic solution; In conventional structure of fuzzy neural network, the process that each fuzzy rule is derived can be expressed as: the linear sum of products of first trying to achieve all input variables in training sample, then with the relevance grade μ of this linear sum of products with rule ⁽ⁱ⁾(X _p) be multiplied, obtain the output of final every bar fuzzy rule; The derivation of fuzzy rule i exports and can be expressed as follows:

$f^{\left(i\right)}={\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})---\left(6\right)$

${\hat{y}}_p=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}{f}^{\left(i\right)}+b=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}[{\mathrm{\μ}}^i\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})]+b---\left(7\right)$

In formula, f ⁽ⁱ⁾be the output of i-th fuzzy rule, that fuzzy neural network model exports the prediction of p training sample, a _ij, j=1 ..., n is the linear coefficient of a jth variable in i-th fuzzy rule, a _i0be the constant term of the linear sum of products of input variable in i-th fuzzy rule, b is output offset amount;

4), in formula (7), the determination of the parameter in the linear sum of products of input variable is the subject matter used during fuzzy neural network uses, here we adopt and original fuzzy rule derivation output form are converted to support vector machine optimization problem, re-use support vector machine and carry out linear optimization, transfer process is as follows:

$\begin{array}{c}{\hat{y}}_p=\underset{i=1}{\overset{R}{\mathrm{\Σ}}}{f}^{\left(i\right)}+b\\ =\underset{i=1}{\overset{R}{\mathrm{\Σ}}}[{\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×(\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×X_{\mathrm{pj}}+a_{i0})]+b\\ =\underset{i=1}{\overset{R}{\mathrm{\Σ}}}\underset{j=0}{\overset{n}{\mathrm{\Σ}}}{a}_{\mathrm{ij}}\×{\mathrm{\μ}}^{\left(i\right)}\left(X_p\right)\×X_{\mathrm{pj}}+b\end{array}---\left(8\right)$

Wherein X _p01 is constantly equal to for constant term; Order

Wherein, represent the reformulations of former training sample, namely original training sample be converted to as above formula form, the training sample as support vector machine:

Wherein y ₁..., y _nbe the target output of training sample, get S as new input training sample set, so original problem can be converted into following support vector machine primal-dual optimization problem:

$R(\mathrm{\ω},b)=\mathrm{\γ}\frac{1}{N}{\mathrm{\Σ}}_{p=1}^N{L}_{\mathrm{\ϵ}}(y_p,f\left(X_p\right))+\frac{1}{2}{\mathrm{\ω}}^T\mathrm{\ω}---\left(11\right)$

Wherein y _pinput training sample X _ptarget export, ω is the normal vector of support vector machine lineoid, f (X _p) be correspond to X _pmodel export, γ is the penalty factor of support vector machine, and R (ω, b) is the objective function of optimization problem, and N is number of training, L _ε(y _p, f (X _p)) expression formula is as follows:

Wherein ε is the error margin of optimization problem, and the forecast of the optimum derivation linear dimensions of fuzzy rule and primal-dual optimization problem that next use support vector machine to try to achieve fuzzy neural network exports:

$a_{\mathrm{ij}}=\underset{k=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\α}}_k^{*}-{\mathrm{\α}}_k){\mathrm{\μ}}^{\left(i\right)}{X}_{\mathrm{kj}}=\underset{k\∈\mathrm{SV}}{\overset{N}{\mathrm{\Σ}}}({\mathrm{\α}}_k^{*}-{\mathrm{\α}}_k){\mathrm{\μ}}^{\left(i\right)}{X}_{\mathrm{kj}},i=1,...,R;j=0,...,n---\left(13\right)$

Wherein α _k, (k=1 ..., N) and be y respectively _p-f (X _p) be greater than 0 and be less than 0 time corresponding Lagrange multiplier, be the training sample X after corresponding to p standardization _pmI predicted value;

Described flexible measurement method is further comprising the steps of: 5), regularly off-line analysis data is input to training sample and concentrates, and upgrades fuzzy neural network model.