CN1414070A - Wax yield on line optimization control system of benzol acetone dewaxing device by corelated integration method - Google Patents

Abstract

An in-line optimized control method for the wax recovering rate in the acetone-benzol dewaxing process includes such steps as reading the temp. flow, etc from basic DSC controller, using correlation integration method to calculate the correlative integration between optimized variables of each channel and the correlative integration between variable and target function (wax recovering rate) per minute, periodically calculating the optimized values of variables, and senting them to said basic controller for execution. Its advantages are high adaptivity and anti-interference power, high control stability, and high wax recovering rate.

Description

The benzol acetone dewaxing device by corelated integration method Wax yield on line optimization control system

The present invention relates to the on-line optimizing and controlling method of ketone-benzol dewaxing process wax yield, belong to the advanced control techniques of process field.

The present invention adopts correlation integral method to the on-line optimization control of ketone-benzol dewaxing device wax yield, has realized the closed-loop optimization control to the dewaxed oil yield.

The benzol-kentone dewaxing deoiler is an important component part of lubricating oil processing in the Crude Oil Processing.The technical process of its production process is long, and operation is many, and raw material type is many.The quality of its operational stage and the benefit equipment situation numerous with it reaches at that time, and the raw material composition has confidential relation.How, whole device is implemented Optimizing operation exactly according at that time operating conditions, improve the pressed distillate or the wax that deoils yield, reduce cost, the economic benefit that improves device is had great significance.

The object of the present invention is to provide a kind of pre-thinning ratio with each stream with device, a solvent ratio (1), a solvent ratio (2), secondary solvent, a plurality of variablees such as cold wash ratio are the closed-loop optimization Controlling System of tuning variable.

The factor that the optimizing process of benzol-kentone dewaxing degreasing process relates to is a lot, and optimization aim can comprise the yield of pressed oil, all many-sides such as yield, treatment capacity, solvent consumption and energy expenditure of wax.The factor that influences the target product yield in solvent dewaxing wax system is a lot, as cold-point temperature, and solvent ratio, filter efficient, raw materials quality or the like all may affect yield.In these numerous influence factors, the crystalline state of wax plays a part crucial.The advantages of good crystallization of wax, oil-containing is few in the wax, and filtration efficiency is also high, and is also helpful to the processing of aftertreatment.The crystalline quality has substantial connection with raw materials quality, cold-point temperature, solvent ratio etc. again.Because raw materials quality is a uncontrollable variable in the dewaxing degreasing process, be difficult to again measure, therefore can not be as the tuning variable.And solvent ratio also has confidential relation to crystallisation process, easy Be Controlled again, therefore optional its as regulating measure.Through testing for many years and studying, we have developed a solvent ratio of overlapping with each stream in the system and have been the tuning means, are the line closed loop optimal control system of objective function with the dewaxed oil yield.This system is based on the online device operating mode that records, and the utilization correlation integral optimization method of exploitation recently is means, and the solvent ratio of each stream in the system is carried out online optimization.

The ultimate principle of correlation integral method on-line optimization Controlling System is: in the on-line optimization Controlling System of flow process type production equipment, use correlation-integral method with the variate-value calculation optimization variable of real-time measurement and the cross-correlation integral of target variable, and obtain the gradient of objective function, and then calculate the new set-point of tuning variable, implement to regulate by the basic controller of device, thereby make objective function be issued to maximum satisfying constraint condition, realize real-time optimization control production equipment.

The gordian technique of correlation integral method on-line optimization Controlling System is the correlation integral theory.Correlation integral is a kind of relevant computing with stochastic process, and in the correlation integral law theory, objective function, interference, optimization variable are regarded as stochastic process, and optimization variable is that average is controlled, usually can be expressed as: $\tilde{J}\left(t\right)=f(\tilde{u}\left(t\right),\tilde{p}\left(t\right),t)$ Wherein Be the controlled tuning variable of m dimension average,

For disturbing, f is a unknown mappings.The optimal objective function definition is: ${\tilde{J}}^{*}\left(t\right)=\underset{E\left\{\tilde{u}\left(t\right)\right\}}{\max }E\left\{f\right[E\left\{\tilde{u}\left(t\right)\right\},\tilde{p}\left(t\right),t\left]\right\}$ Here, Be the average of tuning variable, it can be the set(ting)value of basic unit's controller or valve position etc.For univariate optimization problem, can prove that the objective function average is to the gradient of tuning variable average under certain condition

Satisfy following formula: $K_{\mathrm{uJ}}=K_{\mathrm{uu}}\frac{\mathrm{dEf}}{\mathrm{dE}\left\{\tilde{u}\left(t\right)\right\}}+\mathrm{\&epsiv;}\left(t\right)$ In the formula, ε (t) is that average is zero noise item, and K _UJBe the cross-correlation integral between tuning variable and objective function, K _UuBe tuning variable autocorrelation integral, as follows: $K_{\mathrm{uJ}}(T,M)={\&Integral;}_{-\mathrm{<figure-callout\; id="1"\; label="M\; M"\; filenames="A0213076900111.png"\; state="\{\{state\}\}">M</figure-callout>}}^M\frac{1}{2T}{\&Integral;}_{-T}^{T}u(t-\mathrm{\&tau;})J\left(t\right)\mathrm{dtd\&tau;}$ $K_{\mathrm{uu}}(T,M)={\&Integral;}_{-\mathrm{<figure-callout\; id="1"\; label="M\; M"\; filenames="A0213076900111.png"\; state="\{\{state\}\}">M</figure-callout>}}^M\frac{1}{2T}{\&Integral;}_{-T}^{T}u(t-\mathrm{\&tau;})u\left(t\right)\mathrm{dtd\&tau;}$

T, M＞0 is a constant

In the above-mentioned formula, u (t), J (t) is respectively the observed value of tuning variable and objective function.As seen K _UJ, K _UuCan calculate by the observed value of tuning variable and objective function, so according to formula $K_{\mathrm{uJ}}=K_{\mathrm{uu}}\frac{\mathrm{dEf}}{\mathrm{dE}\left\{\tilde{u}\left(t\right)\right\}}+\mathrm{\&epsiv;}\left(t\right)$

The gradient of available method of least squares estimation objective function

Calculate after the gradient of objective function, can be with the new settings value u of direct iterative computation tuning variable _s(l+1) $u_s(l+1)=u_s\left(l\right)+a\frac{\mathrm{dEf}}{\mathrm{dE}\left\{\tilde{u}\left(l\right)\right\}}$ In the formula, a is a constant.

Online the continuing of this iterative process carries out, and is zero until gradient.For multivariable optimization problem, theoretical analysis also can obtain similar result.

Closed-loop optimization Controlling System of the present invention is the secondary control mode.Optimal controller from the DCS basic controller of benzol-kentone dewaxing production equipment, read must metrical information, adopt correlation-integral method, per minute calculates the once correlation integral between each variable, and the correlation integral between each variable and objective function.Optimal controller is according to these correlation integral value, and the set-point that variable is once regulated and control in calculating regularly is sent to the basic controller execution.So move in circles, realize online closed-loop optimization control, improved device automation of operation degree, do not need the operative employee to intervene substantially, operation and maintenance is more prone to carry out.Because optimization system itself has fault diagnosis system and a series of safety precautions, situation in addition occurs simultaneously, all can switch to conventional control, ensure safety as the malfunctioning grade of instrument so work as device.Fig. 1 is an optimal control structure iron of the present invention.Fig. 2 is the specific embodiments of the invention structured flowchart.

The present invention will be further described below in conjunction with specific embodiment and control process thereof.The technical process of Fushun oil one factory's newly developed area ketone-benzol dewaxing device as shown in Figure 3.

Fushun one factory newly developed area benzol-kentone dewaxing deoiler comprises: dewaxing, one section de-oiling, two sections de-oilings are filtered; Filtrate, wax liquid, sweat oil solvent recuperation and drying and dehydrating; Five systems of nitrogen circulation and ammonia refrigeration and cooling.Because the ketone-benzol dewaxing device Wax yield on line optimization control system relates generally to the crystallization filtering system, below the just main technical process that these two systems are discussed.Stock oil is divided into 7 the tunnel after entering system, every road is provided with 3 crystallizers, the filtrate after interchanger E104 heating from one section wax deoiling is at first added to dilute in advance to stock oil in each road, enter E101 afterwards, when stock oil enters E101, adding is changed filtrate after cold to carry out once (1) dilution from one section wax deoiling through E105 and the fresh solvent that is used for carrying out secondary dilution, stock oil through once (1) dilution and through E101 with change from the filtrate after the dewaxing cold after, behind interchanger E101, add directly and dilute to carry out once (2) from the filtrate of one section wax deoiling, stock oil enters ammonia cooling crystallizer E102 afterwards, E103, add then through E105, the cooled fresh solvent of E106 and E107 is to carry out secondary dilution, stock oil after the crystallization enters D101, the dewaxing filter, filtrate filtered enters recovery system after E101 and stock oil heat exchange, the wax that filters out enters wax flow container D105, after adding mixes with it from the filtrate of two sections filters of wax deoiling, change through the fresh solvent of E106 and secondary dilution and to enter D106 after cold, one section wax deoiling filter.The filtrate of one section wax deoiling filter enters filtrate tank D110, the part of filtrate as solvent add stock oil with dilute in advance, once (1) and once (2) dilute, a part through E108 and the fresh solvent that is used as one section cold wash of wax deoiling and two sections cold wash change cold after, enter solvent recovering system.The wax that filters out of one section wax deoiling filter enters the wax flow container, after the adding fresh solvent mixes with it, enters two sections wax deoiling filters.The filtrate of two sections wax deoiling filters enters filtrate tank D111, enters dewaxing filter wax flow container D105 then.The wax that two sections wax deoiling filters filter out with enter recovery system after the fresh solvent of adding mixes.

Overall structure

The on-line optimization system of present embodiment adopts computer secondary control mode.Basic unit control is finished direct digital control to solvent flux and solvent ratio by the DCS system.Optimal control computer (host computer) can be used the engineer station of DCS system, the host computer of perhaps joining in addition.Host computer calculates an optimization setting value according to the sigtnal interval certain hour of gathering, and is sent to by output channel on the solvent flux set-point of basic unit's Controlling System.Adopting a great advantage of this mode is when host computer breaks down, can cut off getting in touch of host computer and basic unit's Controlling System easily, directly carry out the control of definite value ratio, can improve the handiness and the reliability of system like this by basic unit's Controlling System.

System forms

This system is made up of one group of optimal control and advanced control subsystem, comprising:

The ketone-benzol dewaxing device Wax yield on line optimization control system

Failure testing and emergent optimal control system

Crystallizer stream pressure advanced control system

Oil product switches and stream wax advanced control system

The real-time yield computing system of system

Operation interface

Objective function and optimization variable

Present embodiment is an objective function with the yield of dewaxed oil, with 7 of the pre-thinning ratios of each stream, (1) 7 of solvent ratio, (2) 7 of solvent ratio, secondary solvent, totally 23 variablees of cold wash ratio is the tuning variable, these variablees of on-line optimization, the adjustment by each road solvent ratio improves crystalline state, improve filtration efficiency, it is maximum that the dewaxed oil yield reaches.

Constraint condition

For optimal control can be carried out in the normal range of operation of device, the constraint of bound need be arranged optimization variable, 23 optimization variable have 47 constraint conditions.These constraint conditions can be provided with by the online foundation of operator device situation at that time.In addition, the total solvent of dewax system is than being an important indicator relevant with energy consumption, and the ability of solvent recovering system also is limited, thus in optimizing process, add total solvent than the upper limit as a constraint condition.

The optimal control core

Optimal control is calculated and adopted correlation integral method, and is little to the interference of operation owing to this method does not need process model building is not needed to add test signal, and can adapt to the variation of raw material, the variation of the system characteristic that scrap build brings automatically.

Optimal control software is installed on the engineer work station of DCS, and software adopts DCI Optimizer for DCS.This software is sophisticated commercial correlation integral control software, embeds in the DCS, need not increase new hardware in addition, and very high reliability is arranged.

Failure testing and emergent treatment system

Because the complicacy of ketone benzene process itself adds numerous on-line optimization variablees, total system relates to a large amount of instrument and complicated computer system.For the assurance system in unusual service condition (as certain closed circuitization of stream wax or stop inspection) and the part instrumental system is damaged, under the situation of maintenance, the operation that still can keep optimization system, system should dispose fault, unusual condition automatic checkout system (being designated hereinafter simply as failure testing) and emergent treatment system.In case after malfunction detection system detected abnormal conditions, except providing corresponding guard signal, the automatic emergency treatment system started automatically, takes corresponding measure according to position of fault and classification.System recovered normal working order automatically after fault disappeared.Optimization system has just possessed certain fault-tolerant ability like this, and can handle automatically without manual intervention under abnormal conditions.

Crystallizer stream pressure advanced control system

So-called stream pressure refers to the pressure of the material inlet section of each stream of crystallization workshop section, and it has reflected the situation that each flowstone wax crystalls adheres in crystallizer, when pressure is big, adheres to seriously, easily causes obstruction.Generally after pressure arrives necessarily greatly, no longer allow crystallizer under little solvent ratio, to work, otherwise cause that easily pressure sharply rises, then the serious consequence of Zu Saiing.In this optimization system, the optimization range of each solvent ratio is subjected to the control of each flowing pressure, promptly is that a kind of agent variable is than lower limit optimal control.Particularly, be exactly that the lower value of each of this road time solvent ratio optimization improves automatically after stream pressure rises, and after this road pressure descended, the optimization lower limit of each time solvent ratio can descend automatically.Agent has a nonlinear funtcional relationship than lower limit and stream pressure, improves the solvent ratio lower value automatically by certain algorithm.

Oil product switches and stream wax advanced control system

Benzol-kentone dewaxing is the processing unit (plant) of raw material variety more than, often carries out the switching of raw material in process of production.Therefore, need handle the optimal control of handoff procedure.When system switch to produce oil product, program will enter changeover program, optimizing process is handled and be put into operation according to certain rule.

In real work, also to solve the problem of stream wax.Along with the increase of working hour, wax adhering in crystallizer is also more and more serious, and the pressure of stream also rises thereupon, has arrived certain the time, must change wax.Therefore, the problem that optimal control system must solution wax be brought.Have special program to handle in the system, the workman only need press a button and get final product.

The accumulation yield calculates

Integrating yield in 8 hours of counting yield, calculation formula is as follows:

In yield calculates, will consider automatically that also the integrating numeral puts 0 situation.

Man-machine interface

Mainly finish demonstration, warning and the operation of online detection data.Man-machine interface is mainly a main operation picture, and many sub-menu option are arranged on this picture, and each submenu comprises many functions again.Its display part comprises many pictures, and with the data on certain Velocity Updating picture, shown real time data has the observed value of each solvent ratio, automatic set-point, automatic---many sampled datas and relevant controlled variable such as manual state, solvent ratio lower value.

In the function part, the workman can realize following online in real time operation by keyboard or mouse: several hand-automatic switchovers at any time arbitrarily in each solvent ratio, the change of manual set-point under the manual state, the intervention of automatic set-point under the auto state, the input processing line is other, changes wax stream number or the like.

Present embodiment puts into operation on Fushun oil one factory newly developed area benzol-kentone dewaxing deoiler, has obtained good effect.In optimizing process

Wax (major product) yield: increase in the optimizing process of 66# wax production decision and be approximately about 0.5～0.9

Approximately increase by 2.02～2.2% in the optimizing process of 58# wax production decision

Oil (auxiliary product) yield: increase in the optimizing process of 66# wax production decision and be approximately about 0.5～0.9

Approximately increase by 0.48～0.8 in the optimizing process of 58# wax production decision

Sweat oil (waste material) yield: reduce about 3%～4% in the optimizing process of 66# wax production decision

In the optimizing process of 58# wax production decision about about 2.24%～2.4%

The oleaginousness of wax and contain dosage and all conform to quality requirements in the optimizing process of each production decision.Energy consumption demonstrates downward trend in the optimizing process of 66# wax, and energy consumption remains in the scope of 60～77 kilograms of mark oil/kg feed material in the optimizing process of 58# wax.

The acquisition process of each measurement signal in the above-mentioned embodiment, optimal controller are realized by distributed control system (DCS) and upper computer thereof.This scheme is applied on the full scale plant, has reached automatic adjustment operational variable, optimization aim function, the purpose of raising device target product yield.According to measuring and calculating, the year economic benefit of this on-line optimization Controlling System is about 3,000,000 yuan.

Claims (2)

1, the benzol acetone dewaxing device by corelated integration method Wax yield on line optimization control system, it is characterized in that: in the Wax yield on line optimization control system of ketone-benzol dewaxing device, use correlation-integral method, yield with the device dewaxed oil is an objective function, pre-thinning ratio with each stream, a solvent ratio (1), a solvent ratio (2), a plurality of variablees such as secondary solvent and cold wash ratio are the tuning variable, whether the objective function of the variate-value calculating device by online real time collecting is in maximization, and regular according to this set-point at each tuning variable of line computation, being sent to basic controller carries out, thereby adjustment by each stream solvent ratio, improve crystalline state, improve filtration efficiency, make the dewaxed oil yield be issued to maximum, realize the real-time optimization control of ketone-benzol dewaxing device wax yield at the state that satisfies constraint condition.

2, method according to claim 1, it is characterized in that: whether the objective function (dewaxed oil productive rate) that is calculated device by the real-time variable value application correlation integral method that installs is in maximization, according to each a plurality of variable such as pre-thinning ratio, a solvent ratio (1), a solvent ratio (2), secondary solvent and cold wash ratio that flow of calculation result setting device, to realize the real-time closed-loop optimal control of these variablees.

Images