CN103076741B - Chemical process non-minimum realizes extended mode space quadric form control method - Google Patents

Abstract

The present invention relates to a kind of chemical process non-minimum and realize extended mode space quadric form control method.The simple control device that current employing is traditional, controling parameters relies on technician's experience completely, and control effects is very undesirable.First the inventive method sets up transport function process model based on chemical process real data, excavates basic process characteristic; Then extended mode space quadric form control loop is set up based on this transport function process model; Finally by the parameter calculating quadratic form controller, process object is implemented quadratic form and control.Technical scheme of the present invention is by the means such as data acquisition, process process, prediction mechanism, data-driven, optimization, establish a kind of chemical process non-minimum and realize extended mode space quadric form control method, utilize the method effectively can improve the precision of control, improve and control smoothness.The inventive method compensate for the deficiency of Traditional control, and effectively facilitates the design of controller, ensures the lifting of control performance.

Description

Chemical process non-minimum realizes extended mode space quadric form control method

Technical field

The invention belongs to technical field of automation, relate to a kind of chemical process non-minimum and realize extended mode space quadric form control method.

Background technology

Chemical process is the important component part of China's process flow industry process, and it is whether effective direct to follow-up process treatment process and reduce full-range energy consumption and be all of great importance that it controls.As an industrial significant subject, the raising of raising to whole economic performance of industrial enterprises of Producing Process of Processing Industry level plays vital effect.For this reason, each main technologic parameters of production run must strictly control.Along with the quality requirements of market to petrochemicals is more and more higher, and the development of production Technology, technological process becomes more complicated, though traditional control method meets certain requirement, is difficult to promote level of control further.Simple process control cannot meet the requirement of control accuracy and stationarity, and product percent of pass is low, and unit efficiency is low, defines the requirement developing into the advanced stage such as complex control, Dynamic matrix control from conventional control.And control substantially to adopt traditional simple control device in current actual industrial, controling parameters relies on technician's experience completely, and production cost is increased, and control effects is very undesirable.China's Chemical Engineering Process Control and optimisation technique relatively backward, energy consumption remains high, and control performance is poor, and automaticity is low, is difficult to the demand adapting to energy-saving and emission-reduction and indirectly environmental protection, this wherein one of direct influence factor be the control program problem of system.

Summary of the invention

Target of the present invention is the deficiency for existing Chemical Processing Systems control technology, provides a kind of chemical process non-minimum to realize extended mode space quadric form control method.The method compensate for the deficiency of traditional control method, and while ensureing to control to have higher precision and stability, the form that also ensures is simple and meet the needs of actual industrial process.

First the inventive method sets up transport function process model based on chemical process real data, excavates basic process characteristic; Then extended mode space quadric form control loop is set up based on this transport function process model; Finally by the parameter calculating quadratic form controller, process object is implemented quadratic form and control.

Technical scheme of the present invention is by the means such as data acquisition, process process, prediction mechanism, data-driven, optimization, establish a kind of chemical process non-minimum and realize extended mode space quadric form control method, utilize the method effectively can improve the precision of control, improve and control smoothness.

The step of the inventive method comprises:

(1) utilize chemical process real data to set up transport function process model, concrete grammar is:

Step (1). the input of operation chemical process makes it have individual Spline smoothing, is exported in real time, by real-time for chemical process output valve by recorder record chemical process response curve convert Dimensionless Form to :

Wherein, that chemical process exports in real time steady-state value.

Step (2). choose two calculation levels, , according to the parameter required for following computing formula calculation of transfer function process model :

Wherein, for the Spline smoothing amplitude of chemical process input.

Step (3). the parameter that step (2) obtains is converted into the transport function process model of Laplce's form:

Wherein, for Laplace transform operator, for the time constant of model, for the time lag of transport function process model, represent the Laplace transform of the output valve of process model, represent the Laplace transform of the input of process model.

(2) realize extended mode space quadric form controller based on this transport function process model design non-minimum, concrete grammar is:

A. above-mentioned transport function process model is passed through the sampling period be converted into discrete input/output model:

Wherein with output and the input variable of discrete input/output model respectively, with be respectively with coefficient polynomial expression;

Wherein corresponding coefficient, for after move step operator, it is the discrete input/output model order obtained;

B. above-mentioned discrete input/output model is passed through backward shift operator be processed into state space form:

。

Wherein, , respectively the state variable in moment and output variable value, be the input incremental variable value in moment, , be respectively the output variable increment in moment and input variable increment size, , , be respectively corresponding state matrix, input matrix and output matrix, for getting transposition symbol.

；

C. defining a process desired output is , and output error for:

Obtain further the output error in moment for:

Wherein, be the process desired output increment in moment.

Define a new combined state variable:

Be comprehensively that a non-minimum realizes extended mode steric course model by above-mentioned processing procedure:

Wherein, be the combined state variable in moment, , , be respectively the state matrix of corresponding combined state variable, input matrix and output matrix, specifically:

， ，

D. defining the vector form that non-minimum realizes extended mode space quadric form controller objective function is:

Wherein, for objective function, with be respectively the weighting matrix of state variable and output variable.

E. the parameter of computing controller, specifically:

Wherein for controller feedback factor vector.

A kind of chemical process non-minimum that the present invention proposes realizes the deficiency that extended mode space quadric form control method compensate for Traditional control, and effectively facilitates the design of controller, ensures the lifting of control performance, meets given production performance index simultaneously.

The control technology that the present invention proposes effectively can reduce the error between desired process parameter and actual process parameter, compensate for the deficiency of traditional controller further, ensure that control device operates in optimum condition simultaneously, make the technological parameter of production run reach strict control.

Embodiment

Control for coking fractional distillation column level process:

Here control to be described as an example with coking fractional distillation column level process.This process is a complicated process, and coking fractional distillation column liquid level is not only subject to the impact of feed rate, simultaneously also by Outlet Temperature in Delayed Coking Furnace, and the impact of fuel quantity flow.Regulating measure adopts feed rate, and remaining impact is as uncertain factor.

(1) set up transport function process model, concrete grammar is:

The first step: the inlet amount valve of operation coking fractional distillation column level process makes it input a Spline smoothing, data acquisition unit is utilized to gather coking fractional distillation column level process input data (feed rate) and export data (coking fractional distillation column liquid level), exported in real time by recorder recording process, by real-time for process output valve response convert Dimensionless Form to :

Wherein, that process exports in real time steady-state value.

Second step: choose two calculation levels, , according to the parameter required for following computing formula calculation of transfer function process model :

Wherein, for the Spline smoothing amplitude of process input.

3rd step: the parameter obtained by second step is converted into the transport function process model of Laplce's form:

Wherein, for Laplace transform operator, for the time constant of model, for the time lag of transport function process model, represent the Laplace transform of the output valve of process model, represent the Laplace transform of the input of process model.

(2) design coking fractional distillation column level process non-minimum and realize extended mode space quadric form controller, concrete grammar is:

A. transport function process model obtained above is passed through the sampling period being converted into discrete input/output model is:

Wherein with output and the input variable of discrete input/output model respectively, with be respectively with coefficient polynomial expression;

Wherein corresponding coefficient, for after move step operator, it is the discrete input/output model order obtained;

B. above-mentioned discrete input/output model is passed through backward shift operator be processed into state space form:

。

Wherein, , respectively the state variable in moment and output variable value, be the input incremental variable value in moment, , be respectively the output variable increment in moment and input variable increment size, , , be respectively corresponding state matrix, input matrix and output matrix, for getting transposition symbol.

；

C. the process that defines expects that liquid level exports , and liquid level output error for:

Obtain further the liquid level output error in moment for:

Wherein, be the process in moment expects liquid level output increment.

Define a new combined state variable:

Be comprehensively that a non-minimum realizes extended mode steric course model by above-mentioned processing procedure:

Wherein, be the combined state variable in moment, , , be respectively the state matrix of corresponding combined state variable, input matrix and output matrix, specifically:

， ，

D. defining the vector form that non-minimum realizes extended mode space quadric form controller objective function is:

Wherein, for objective function, with be respectively the weighting matrix of state variable and output variable.

E. the parameter of computing controller, specifically:

Wherein for controller feedback factor vector.

Claims (1)

1. chemical process non-minimum realizes extended mode space quadric form control method, it is characterized in that the concrete steps of the method are:

I. utilize chemical process real data to set up transport function process model, concrete grammar is:

Step (1): the input of operation chemical process makes it have individual Spline smoothing, is exported in real time, convert the response curve of real-time for chemical process output valve y (k) to Dimensionless Form y by recorder record chemical process ^*(k):

y ^*(k)＝y(k)/y(∞)

Wherein, y (∞) is the steady-state value that chemical process exports y (k) in real time;

Step (2): choose two calculation levels, y ^*(k ₁)=0.39, y ^*(k ₂)=0.63, according to the parameter K required for following computing formula calculation of transfer function process model ₁, T ₁and τ ₁:

K ₁＝y(∞)/q ₁

T ₁＝2(k ₁-k ₂)

τ ₁＝2k ₁-k ₂

Wherein, q ₁for the Spline smoothing amplitude of chemical process input;

Step (3): the transport function process model parameter that step (2) obtains being converted into Laplce's form:

$\frac{y\left(s\right)}{q_1\left(s\right)}=\frac{1}{{\mathrm{\λ}}_{1}s+1}{e}^{-L_{1}s}$

Wherein, s is Laplace transform operator, λ ₁for the time constant of model, L ₁for the time lag of transport function process model, y (s) represents the Laplace transform of the output valve of process model, q ₁s () represents the Laplace transform of the input of process model;

λ ₁＝T ₁

L ₁＝τ ₁

II. realize extended mode space quadric form controller based on this transport function process model design non-minimum, concrete grammar is:

A. above-mentioned transport function process model is passed through sampling period T _sbe converted into discrete input/output model:

F(z ^-1)y(k)＝H(z ^-1)u(k)

Wherein y (k) and u (k) is output and the input variable of discrete input/output model respectively, F (z ^-1) and H (z ^-1) be the coefficient polynomial expression of y (k) and u (k) respectively;

F(z ^-1)＝1+f ₁z ^-1+f ₂z ^-2+…+f _nz ^-n

H(z ^-1)＝h ₁z ^-1+h ₂z ^-2+…+h _nz ^-n

Wherein f _i, h _icorresponding coefficient, i=1,2 ..., n; z ^-ifor after move i walk operator, i=1,2 ..., n; N is the discrete input/output model order obtained;

B. above-mentioned discrete input/output model is processed into state space form by backward shift operator Δ:

Δx _m(k+1)＝A _mΔx _m(k)+B _mΔu(k)

Δy(k+1)＝C _mΔx _m(k+1)

Δx _m(k) ^T＝[Δy(k)Δy(k-1)…Δy(k-n+1)Δu(k-1)Δu(k-2)…Δu(k-n+1)]；

Wherein, Δ x _m(k+1), Δ y (k+1) is state variable and the output variable value in kth+1 moment respectively, the input incremental variable value that Δ u (k) is the kth moment, Δ y (k-i), Δ u (k-i) are respectively output variable increment and the input variable increment size in kth-i moment, i=0,1,, n-1, A _m, B _m, C _mbe respectively corresponding state matrix, input matrix and output matrix, T is for getting transposition symbol;

$A_m=\left[\begin{array}{ccccccccc}-f_1& -f_2& ...& -f_{n-1}& -f_n& h_2& ...& h_{n-1}& h_n\\ 1& 0& ...& 0& 0& 0& ...& 0& 0\\ 0& 1& ...& 0& 0& 0& ...& 0& 0\\ \·& \·& & \·& \·& \·& & \·& \·\\ \·& \·& ...& \·& \·& \·& ...& \·& \·\\ \·& \·& & \·& \·& \·& & \·& \·\\ 0& 0& ...& 1& 0& 0& ...& 0& 0\\ 0& 0& ...& 0& 0& 0& ...& 0& 0\\ 0& 0& ...& 0& 0& 1& ...& 0& 0\\ \·& \·& & \·& \·& & \·& \·& \·\\ \·& \·& ...& \·& \·& ...& \·& \·& \·\\ \·& \·& & \·& \·& & \·& \·& \·\\ 0& 0& ...& 0& 0& 0& ...& 1& 0\end{array}\right]$

B _m＝[h ₁00…0100] ^T

C _m＝[100…0000]；

C. defining a process desired output is r (k), and output error e (k) is:

e(k)＝y(k)-r(k)

The output error e (k+1) obtaining kth+1 moment is further:

e(k+1)＝e(k)+C _mA _mΔx _m(k)+C _mB _mΔu(k)-Δr(k+1)

Wherein, Δ r (k+1) the process desired output increment that is kth+1 moment;

Define new combined state variable z (k):

$z\left(k\right)=\left[\begin{array}{c}\mathrm{\Δ}{x}_m\left(k\right)\\ e\left(k\right)\end{array}\right]$

Be comprehensively that a non-minimum realizes extended mode steric course model by above-mentioned processing procedure:

z(k+1)＝Az(k)+BΔu(k)+CΔr(k+1)

Wherein, the combined state variable that z (k+1) is kth+1 moment, A, B, C are respectively the state matrix of corresponding combined state variable, input matrix and output matrix, specifically:

$A=\left[\begin{array}{cc}{A}_m& 0\\ C_m{A}_m& 1\end{array}\right],B=\left[\begin{array}{c}{B}_m\\ C_m{B}_m\end{array}\right],C=\left[\begin{array}{c}0\\ -1\end{array}\right]$

D. defining the vector form that non-minimum realizes extended mode space quadric form controller objective function is:

$J=\underset{k=1}{\overset{\∞}{\mathrm{\Σ}}}[z{\left(k\right)}^T\mathrm{Qz}\left(k\right)+\mathrm{\λ\Δu}{\left(k\right)}^2]$

Wherein, J is objective function, Q and λ is respectively the weighting matrix of state variable and output variable;

E. the parameter of computing controller, Δ u (k)=-Kz (k), wherein K is controller feedback factor vector.