CN102553281A - Optimization method of fractionation system in adsorption separation device - Google Patents

Abstract

The invention discloses an optimization method of a fractionation system in an adsorption separation device, which comprises the steps of: 1) collecting the real-time and historical operation data of the fractionation system in the adsorption separation device, wherein the real-time and historical operation data comprise composition analysis and flow rate of mixed C8A entering the adsorption separation device; composition analysis, temperature, pressure and flow rate of liquid extract entering the fractionation system; composition analysis, temperature, pressure and flow rate of raffinate entering the fractionation system; and composition analysis of tower top products of a raffinate tower and an extract tower in the fractionation system, tower kettle product composition, tower top reflux ratio and temperature distribution in the tower; and 2) determining the optimization operation variables and constraint conditions of the fractionation system by utilizing the concentrations of the key impurities in the tower top and tower kettle products of the fractionating towers as the constraint conditions, and by aiming at obtaining tower top cold reflux of the extract tower and the raffinate tower, temperature distribution in the tower, and the minimal error between the reboiling load of the tower kettle and the actual operation value of the device. The optimization method has the advantage of solving the problem that the key technique parameters of the current fractionation systems cannot be correspondingly adjusted with the changes of the mixed C8A composition and the feeding load.

Description

The optimization method of fractionating system in the adsorption separation device

One, technical field

The invention belongs to the separating technology in the Chemical Engineering, relate to the optimization method of fractional distillation process in the mixed C 8 aromatic hydrocarbons adsorptive separation technologies, especially relate to the optimization of raffinate and Extract tower sharing of load and operating parameter.

Two, background technology

Aromatic hydrocarbons is one of most important basic organic material of petro chemical industry.In Chinese national economy and social development, play an important role.Four kinds of isomers of C8 aronmatic are respectively that paraxylene (PX), ortho-xylene (OX), meta-xylene (MX) and ethylbenzene (EB) generally all are that form with mixture exists, and it is a bigger link of difficulty in the PX production technology that mixed C 8A (PX, MX, OX, EB) separates.Separate in the commercial run of four kinds of isomers of mixed C 8A, mainly contain cryogenic crystallization method, Separation by Complexation method, adsorption method of separation, secondly also have methods such as eutectic, sulfonation.Based on the simulation moving-bed molecular sieve adsorption isolation technics of continuous flow upstream is the main flow technology of producing PX at present, also has minority to adopt crystallization (deep cooling) technology.It is a new isolation technics that is at first proposed the sixties in last century and developed by American UOP company that moving-bed adsorption is separated; Should the technology representative processes be the Eluxyl technology of the Parex technology of UOP (UOP), French IFP (Compagnie Francise des Petroles) and the Aromax technology of Japanese Toray (east is beautiful) company at present; This technology combines with isomerization process usually; Produce PX with high yield; Generally adopted by various countries at present, wherein American UOP company and French Inst Francais Du Petrole have occupied the market share near 90%, occupy monopoly position.Continuous development and growth along with China's chemical industry; Commercial production is becoming tight to continuous growth of the demand of basic organic materials such as aromatic hydrocarbons and international petroleum supply day by day; Improve the technological level of aromatic hydrocarbons production process, holding device is efficient, even running becomes the leading indicator that domestic each aromatic hydrocarbons manufacturer promotes enterprise's international competitiveness and weighs the device technique level.

Adsorption separation device mainly is to utilize the difference of adsorbent to four kinds of isomers adsorption capacities of mixed C 8A; Wherein to the characteristic of PX preferential adsorption; The PX among the mixed C 8A preferentially adsorbs in the adsorption tower with getting into continuously, and through refining and resolve, the mixed C 8A that gets into adsorption tower the most at last separates two strands of materials of journey simultaneously: resolve the mixed liquor of agent and thick PX and resolve agent and contain the C8A mixed liquor of poor PX; Obtain highly purified PX product through rectifying then and resolve agent with C8A that contains poor PX and circulation; The C8A of poor PX delivers to isomerization unit, resolves agent and recycles, and highly purified PX sees off as product.The adsorption separation system of existing aromatic hydrocarbons production technology generally all is based on certain or several typical production status carry out technological design and lectotype selection; And load and performance of the adsorbent difference owing to mixed C 8A composition, apparatus processes in the actual production process, make that the technological parameter off-design parameter of adsorption separation system is obvious.Especially as far as having the device of many parallelly connected production lines; Its performance of the adsorbent difference is self-evident, and mixed C 8A forms and the apparatus processes load all can disturb the efficiently and smoothly operated generation of key equipments such as the Extract tower in the follow-up fractionating system, raffinate tower.Therefore form based on C8A and change and performance of the adsorbent difference; Crucial production processes such as the Extract tower of adsorption separation device, the rectifying of raffinate tower are optimized method research; Help the efficiently and smoothly operated of fractionating system, reduce plant energy consumption and resolve agent consumption.

Therefore, the optimization method of the adsorption separation system of C8 aronmatic---the Extract, raffinate tower sharing of load and the operation parameter optimization that especially relate to performance of the adsorbent difference can have influence on commercial production efficient greatly.

Three, summary of the invention

The objective of the invention is: to have the weak point that arene adsorptive separation apparatus exists now in actual motion; A kind of optimization method of fractionating system technology operational factor simple to operate, easy to implement has been proposed; Success solves because of mixed C 8A is the adverse effect that C8 aronmatic composition, processing load variations and performance of the adsorbent difference cause fractionating system in the adsorption separation device, to reduce plant energy consumption and the stable operation of resolving agent consumption and assurance adsorption separation device.

Technical scheme of the present invention is: the optimization method of fractionating system operational factor in a kind of adsorption separation device.It is characterized in that this method may further comprise the steps:

1) at first gather fractionating system in the adsorption separation device in real time and history data; The service data that needs collection: comprise the mixed C 8A composition analysis and the flow that get into adsorption separation device; Get into Extract composition analysis, temperature, pressure and the flow of fractionating system; Get into raffinate composition analysis, temperature, pressure and the flow of fractionating system; Raffinate column overhead product composition analysis, tower still product are formed on every production line in the fractionating system, overhead reflux is than Temperature Distribution in, the tower; Temperature Distribution in Extract column overhead product composition analysis, tower still product composition, overhead reflux ratio, the tower on every production line;

2) adopt data coordinating method that relevant service data is in harmonious proportion, guarantee that the device service data of being gathered is true and reliable;

(1) definite not survey variable U that surveys variable X and need estimate that need coordinate, and by the variable X of having measured _iSample number estimate the variance of measure error; X _iFor m ties up i dimension variable in the tested vector;

(2) preset the ratio η that human error occurs and the ratio γ of human error standard deviation and random error standard deviation in error;

(3) adopt optimized Algorithm to find the solution object function:

$\min -\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\ln \left\{(1-\mathrm{\η})\exp \right[-\frac{1}{2}\frac{{(x_i-{\hat{x}}_i)}^2}{{\mathrm{\σ}}_i^2}]+\frac{\mathrm{\η}}{\mathrm{\γ}}\exp [-\frac{1}{2}\frac{{(x_i-{\hat{x}}_i)}^2}{{\mathrm{\γ}}^2{\mathrm{\σ}}_i^2}\left]\right\}$

$s.t.F(\hat{X},\hat{U})=0$

Wherein m is the dimension of tested vector, x _iFor m ties up the measured value of the dimension of i in tested vector variable,

For m ties up the actual value of the dimension of i in tested vector variable, For obeying the normal distribution variance of instrument measurement error, X ^LAnd X ^UBe the bound of measurand, U ^LAnd U ^UBe the bound of measurand not.

(4) error

of calculating each measurand judges whether to exist human error through judgment criterion ; If unerring error exists, return

algorithm and finish; Otherwise continue.

(5) judgement is existed the e of human error _i, calculate:

Human error ξ with maximum _iThe process measurement variable that is reacted obtains new X and U as not surveying variable, returns (3).

In the above-mentioned algorithm possible human error has been taked the strategy of rejecting one by one, thereby reduced the influence of human error data coordination result;

3) the device service data is being carried out on the basis of dynamic coordinate, according to the component materials equilibrium equation of stable state chemical process, the equation that balances each other, and enthalpy balance equation etc. set up the mechanism model of fractionating system;

4) the combined with intelligent optimized Algorithm is carried out corrected model parameter to the mechanism model of fractionating system, obtains reflecting the industry pattern of commercial plant actual operating mode; The industry pattern of said foundation is a constraints with critical impurities concentration in each fractionator overhead, the tower still product; Temperature Distribution, tower still in each the Extract tower that is obtained by Model Calculation, raffinate column overhead cold reflux, the tower load that boils again is an object function with device actual motion value deviation minimum, that is:

$\min (\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{\left(\frac{Q_i-Q_i^{\%}}{Q_i^{\%}}\right)}^2+\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\left(\frac{T_j-T_j^{\%}}{T_j^{\%}}\right)}^2+{\left(\frac{R-R^{\%}}{R^{\%}}\right)}^2)$

Wherein tower still thermic load, the n of each rectifying column is that the total number of temperature point, the R of each rectifying column is demethanizer column overhead methane cold reflux amount in the fractionating system in the m fractionating system;

5) confirm the Optimizing operation variable and the constraints of fractionating system, set up the fractionating system energy consumption, resolve agent consumption and the mixed C that gets into adsorption separation device ₈Experience database between A composition, apparatus processes load and the performance of the adsorbent; The multiple target distribution estimating optimized Algorithm of using Density Estimator based on above-mentioned industry pattern is to certain mixed C ₈Fractionating system operational factor in the adsorption separation device under A forms and the total processing of device is loaded is optimized to be found the solution, and Considering Adsorption agent performance difference carries out the optimization of matching that fractionating system is handled load to the collaborative influence of its follow-up piece-rate system simultaneously.

The present invention is based on industry pattern; The multiple target distribution estimating algorithm of application Density Estimator is optimized the fractionating system operational factor in the adsorption separation device with the optimized Algorithm that experience database merges mutually and finds the solution; Considering Adsorption agent performance difference carries out the optimization of matching that fractionating system is handled load to the collaborative influence of its follow-up piece-rate system simultaneously, reaches to reduce adsorption separation device energy consumption and the purpose of resolving agent consumption, device even running as much as possible.

The invention has the beneficial effects as follows: optimization method of the present invention can be formed according to the mixed C 8A that gets into adsorption separation device, handle load and performance of the adsorbent difference, and the fractionating system technology operational factor in the adsorption separation device is optimized configuration.Especially form, handle load at mixed C 8A and depart from the Design of device parameter when obvious, by process parameter optimizing significantly steadily mixed C 8A form and handle the fluctuation that cause of load variations fractionating system, reduce plant energy consumption and material consumption simultaneously.The present invention is a kind of optimization method of fractionating system technology operational factor simple to operate, easy to implement; The success solution is the adverse effect that C8 aronmatic composition, processing load variations and performance of the adsorbent difference cause fractionating system in the adsorption separation device because of mixed C 8A; Guarantee the stable operation of adsorption separation device, and in production reality, use.

The specific embodiment

Embodiment: operating parameters such as the feed loading of adsorption separation system, reflux ratio are optimized guidance by optimization method of the present invention; Make Extract tower and raffinate tower PDEB loss reduce more than 20%; Be that PDEB concentration has been reduced to 40ppm by 50ppm in the product, the also corresponding increase of smoothness of Extract tower, the operation of raffinate tower; Install comprehensive energy consumption simultaneously and reduce 4kg mark oil/ton .PX.

Process in accordance with the present invention can be applied to produce actual,

The service data that said step 1 needs to gather comprises the mixed C that gets into adsorption separation device ₈A composition analysis and flow; Get into Extract composition analysis, temperature, pressure and the flow of fractionating system; Get into raffinate composition analysis, temperature, pressure and the flow of fractionating system; Raffinate column overhead product composition analysis, tower still product are formed on every production line in the fractionating system, overhead reflux is than Temperature Distribution in, the tower; Temperature Distribution in Extract column overhead product composition analysis, tower still product composition, overhead reflux ratio, the tower on every production line;

The algorithm steps that data are coordinated in the said step 2 is:

Definite needs are coordinated surveys variable X and the not survey variable U that need estimate, and estimates the variance of measure error by the measurand sample.

(1) presets the ratio η that human error occurs and the ratio γ of human error standard deviation and random error standard deviation in error.

(2) adopt optimized Algorithm to find the solution object function:

$\min -\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\ln \left\{(1-\mathrm{\η})\exp \right[-\frac{1}{2}\frac{{(x_i-{\hat{x}}_i)}^2}{{\mathrm{\σ}}_i^2}]+\frac{\mathrm{\η}}{\mathrm{\γ}}\exp [-\frac{1}{2}\frac{{(x_i-{\hat{x}}_i)}^2}{{\mathrm{\γ}}^2{\mathrm{\σ}}_i^2}\left]\right\}$

$s.t.F(\hat{X},\hat{U})=0$

$X^L\≤\hat{X}\≤X^U;U^L\≤\hat{U}\≤U^U$

(3) error

of calculating each measurand judges whether to exist human error through judgment criterion

; If unerring error exists, return

algorithm and finish; Otherwise continue.

(4) judgement is existed the e of human error _i, calculate:

ξ with maximum _iThe process measurement variable that is reacted obtains new X and U as not surveying variable, returns (3).

In the above-mentioned algorithm possible human error has been taked the strategy of rejecting one by one, thereby reduced the influence of human error data coordination result.

The industry pattern of setting up in the said step 4 is a constraints with critical impurities concentration in each fractionator overhead, the tower still product; Temperature Distribution, tower still in each the Extract tower that is obtained by Model Calculation, raffinate column overhead cold reflux, the tower load that boils again is an object function with device actual motion value deviation minimum, that is:

$\min (\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{\left(\frac{Q_i-Q_i^{\%}}{Q_i^{\%}}\right)}^2+\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\left(\frac{T_j-T_j^{\%}}{T_j^{\%}}\right)}^2+{\left(\frac{R-R^{\%}}{R^{\%}}\right)}^2)$

Wherein tower still thermic load, the n of each rectifying column is that the total number of temperature point, the R of each rectifying column is demethanizer column overhead methane cold reflux amount in the fractionating system in the m fractionating system.

In the said step 5 based on industry pattern; The multiple target distribution estimating algorithm of application Density Estimator is optimized the fractionating system operational factor in the adsorption separation device with the optimized Algorithm that expertise merges mutually and finds the solution; Considering Adsorption agent performance difference carries out the optimization of matching that fractionating system is handled load to the collaborative influence of its follow-up piece-rate system simultaneously, reaches to reduce adsorption separation device energy consumption and the purpose of resolving agent consumption, device even running as much as possible.

Said industry pattern is in that the device service data is carried out on the basis of dynamic coordinate, according to the component of stable state chemical process

Material balance equation, the equation that balances each other, and enthalpy balance equation etc. set up the mechanism model of fractionating system, promptly

The component materials equilibrium equation:

Equation balances each other:

$\left\{\begin{array}{c}{Y}_{i,n}=K_{i,n}{X}_{i,n}\\ i=1,...,N_c\\ n=1,...,N_s\end{array}\right.$

The enthalpy balance equation:

$\left\{\begin{array}{c}-H_{\mathrm{Ln}-1}{L}_{n-1}^{\%}+H_{\mathrm{Ln}}{L}_n+H_{\mathrm{Vn}}{V}_n-H_{\mathrm{Vn}+1}{V}_{n+1}^{\%}=H_{\mathrm{Fn}}{F}_n+Q_n\\ n=1,...,N_s\end{array}\right.$

Molar fraction normalizing equation:

$\left\{\begin{array}{c}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{\mathrm{ij}}=1\\ \underset{i=1}{\overset{N}{\mathrm{\Σ}}}{X}_{\mathrm{ij}}=1\\ j=\mathrm{1,2,3},...,N_S\end{array}\right.$

In the above equation, i is the component mark, and n is the column plate mark, and N is total number of components, and NS is total number of plates, and K is a phase equilibrium constant, and X is that liquid phase is formed mole fraction, and Y is that vapour phase is formed mole fraction, F _nBe inlet amount, Q _nBe heat, L _nFor leaving the amount of liquid of n layer column plate, L _N+1Be the amount of liquid from the upper strata, V _N+1Be the gas flow from lower floor, V _nFor leaving the gas flow of n layer column plate, HL is the liquid phase enthalpy, and HV is the vapour phase enthalpy.

Claims (2)

1. the optimization method of fractionating system operational factor in the adsorption separation device is characterized in that may further comprise the steps:

1) at first gather fractionating system in the adsorption separation device in real time and history data; The service data that needs collection: comprise the mixed C 8A composition analysis and the flow that get into adsorption separation device; Get into Extract composition analysis, temperature, pressure and the flow of fractionating system; Get into raffinate composition analysis, temperature, pressure and the flow of fractionating system; Raffinate column overhead product composition analysis, tower still product are formed on every production line in the fractionating system, overhead reflux is than Temperature Distribution in, the tower; Temperature Distribution in Extract column overhead product composition analysis, tower still product composition, overhead reflux ratio, the tower on every production line;

2) adopt data coordinating method that relevant service data is in harmonious proportion, guarantee that the device service data of being gathered is true and reliable;

(1) definite not survey variable U that surveys variable X and need estimate that need coordinate, and by the variable X of having measured _iSample number estimate the variance of measure error; X _iFor m ties up i dimension variable in the tested vector;

(2) preset the ratio η that human error occurs and the ratio γ of human error standard deviation and random error standard deviation in error;

(3) adopt optimized Algorithm to find the solution object function:

$\min -\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\ln \left\{(1-\mathrm{\η})\exp \right[-\frac{1}{2}\frac{{(x_i-{\hat{x}}_i)}^2}{{\mathrm{\σ}}_i^2}]+\frac{\mathrm{\η}}{\mathrm{\γ}}\exp [-\frac{1}{2}\frac{{(x_i-{\hat{x}}_i)}^2}{{\mathrm{\γ}}^2{\mathrm{\σ}}_i^2}\left]\right\}$

$s.t.F(\hat{X},\hat{U})=0$

Wherein m is the dimension of tested vector, x _iFor m ties up the measured value of the dimension of i in tested vector variable,

For m ties up the actual value of the dimension of i in tested vector variable,

For obeying the normal distribution variance of instrument measurement error, X ^LAnd X ^UBe the bound of measurand, U ^LAnd U ^UBe the bound of measurand not;

(4) error

of calculating each measurand judges whether to exist human error through judgment criterion

; If unerring error exists, return algorithm and finish; Otherwise continue;

(5) judgement is existed the e of human error _i, calculate:

Human error ξ with maximum _iThe process measurement variable that is reacted obtains new X and U as not surveying variable, returns (3);

In the above-mentioned algorithm possible human error has been taked the strategy of rejecting one by one, thereby reduced the influence of human error data coordination result;

With critical impurities concentration in each fractionator overhead, the tower still product is constraints, and Temperature Distribution, the tower still in each the Extract tower that obtains, raffinate column overhead cold reflux, the tower boils to load and install actual motion value deviation minimum again and be object function, that is:

$\min (\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{\left(\frac{Q_i-Q_i^{\%}}{Q_i^{\%}}\right)}^2+\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\left(\frac{T_j-T_j^{\%}}{T_j^{\%}}\right)}^2+{\left(\frac{R-R^{\%}}{R^{\%}}\right)}^2)$

Wherein tower still thermic load, the n of each rectifying column is that the total number of temperature point, the R of each rectifying column is demethanizer column overhead methane cold reflux amount in the fractionating system in the m fractionating system;

Confirm the Optimizing operation variable and the constraints of fractionating system, set up the fractionating system energy consumption, resolve agent consumption and the mixed C that gets into adsorption separation device ₈Experience database between A composition, apparatus processes load and the performance of the adsorbent; The multiple target distribution estimating optimized Algorithm of using Density Estimator based on above-mentioned mechanism model is to certain mixed C ₈Fractionating system operational factor in the adsorption separation device under A forms and the total processing of device is loaded is optimized to be found the solution, and Considering Adsorption agent performance difference carries out the optimization of matching that fractionating system is handled load to the collaborative influence of its follow-up piece-rate system simultaneously.

2. the optimization method of fractionating system operational factor in the adsorption separation device according to claim 1, it is characterized in that according to material balance equation, the equation that balances each other, and enthalpy balance equation etc. set up the mechanism model of fractionating system, promptly

The component materials equilibrium equation:

Equation balances each other:

$\left\{\begin{array}{c}{Y}_{i,n}=K_{i,n}{X}_{i,n}\\ i=1,...,N_c\\ n=1,...,N_s\end{array}\right.$

The enthalpy balance equation:

$\left\{\begin{array}{c}-H_{\mathrm{Ln}-1}{L}_{n-1}^{\%}+H_{\mathrm{Ln}}{L}_n+H_{\mathrm{Vn}}{V}_n-H_{\mathrm{Vn}+1}{V}_{n+1}^{\%}=H_{\mathrm{Fn}}{F}_n+Q_n\\ n=1,...,N_s\end{array}\right.$

Molar fraction normalizing equation:

$\left\{\begin{array}{c}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{Y}_{\mathrm{ij}}=1\\ \underset{i=1}{\overset{N}{\mathrm{\Σ}}}{X}_{\mathrm{ij}}=1\\ j=\mathrm{1,2,3},...,N_S\end{array}\right.$

In the above equation, i is the component mark, and n is the column plate mark, and N is total number of components, and NS is total number of plates, and K is a phase equilibrium constant, and X is that liquid phase is formed mole fraction, and Y is that vapour phase is formed mole fraction, F _nBe inlet amount, Q _nBe heat, L _nFor leaving the amount of liquid of n layer column plate, L _N+1Be the amount of liquid from the upper strata, V _N+1Be the gas flow from lower floor, V _nFor leaving the gas flow of n layer column plate, HL is the liquid phase enthalpy, and HV is the vapour phase enthalpy.