CN100396849C - Controlled addition of co2 in the wet end of the papermaking process - Google Patents

Abstract

The invention provides a method for control added CO2 in a wet-part processing of adding CO2, which includes combining the paper making composition and CO2 to produce a paper making composition containing CO2, determining or estimating at least one electricity capacity of the said paper making composition or paper making composition containing CO2, and controlling the adding speed of CO2 to retain at least one electricity capacity in the preelection range.

Description

CO 2Controlled interpolation in the paper technology wet end

Background of invention

Invention field

The present invention relates to adopting interpolation CO ₂Wet end processing in control CO ₂The method of adding.

Association area

The invention discloses a kind of at employing interpolation CO ₂Wet end processing in control CO ₂The method of adding.The application's assignee, Air Liquide has one at present about CO in the wet end processing ₂The unexamined patent application of adding.

In sheet processing, because competition in the industry, raw materials variations, environmental consideration factor and bigger customer demand amount, understanding and the understanding higher level to wet end chemistry are important for commercial achieving success.Have been found that zeta potential (ZP) is an important performance that constitutes final paper quality.

People early have recognized that, zeta potential is because relevant with flocculation, delay and the drainage performance of paper pulp, thereby are the important factor in order of paper machine process operability.Referring to N.Vanderhoek, " measure the performance (Optimizing Paper MachinePerformance Through Electrokinetic Measurement) of optimizing paper machine by electrokinetics " (APPITA, the 47th volume, the 5th phase, the 397-405 page or leaf, 1994).Wet end stability is the key component that obtains machine efficiency.T.Miyanashi etc. have studied the influence of zeta potential to these phenomenons, and draw and should control zeta potential better to be flocculated and the conclusion of draining additive.The wet end chemistry of optimization paper machine is not operated under equilibrium condition, thereby the interpolation rule of chemicals is very crucial.Referring to T.Miyanashi and S.Motegi, " by the flocculation and the draining (Optimizing Flocculation and Drainage for Microparticleby Controlling Zeta Potential) of control zeta potential optimization particulate " (TAPPI periodical, the 262-270 page or leaf, in January, 1997).

The paper machine of the better operating characteristics of needs has triggered the research and development of better retention agent.This has determined to understand these novel wet end additives better, and can control its interpolation in real time so that obtain best paper machine and whole wet end operation performance.Realize these targets, the first step depends on obtainable real time information.Therefore be desirable to provide a kind of method, it can regulate self to adapt to real-time variation.In wet end technology, the character of fiber characteristics, chemical addition agent and water can change continuously, therefore wishes to detect these variations when it takes place in real time.

As mentioned above, Air Liquide has researched and developed and a kind ofly can utilize CO ₂Change the technology of cellulose fibre zeta potential.M.Muguet and J.M.deRigaurd, " improve the manufacturing technique (Improvements to Processesfor Manufacturing Paper Products by Improving thePhysico-Chemical Behavior of the Paper Stock) of paper products by the physical and chemical performance that improves paper making raw material ", international publication number No.WO 03/074788A2.Air Liquide is in this technical patent application that a pending trial is also arranged, and No. 6052 applications that on September 30th, 2002 submitted as U.S. Provisional Application obtain sequence number 60/414,876, and become U.S.'s non-provisional application on September 6th, 2003, obtain sequence number 10/656,857.The applicant is incorporated herein by reference the full text of these disclosures here clearly, has just done complete elaboration in this application as them.

In process control, need a Mathematical Modeling that the parameter and the output variable of control system of input is interrelated usually.In order to control the wet end parameter, for example delay of wet end additive, discharging and formation at first needs a Mathematical Modeling that the additive parameter is associated with zeta potential and cationic demand.Then, need a model, it will influence the parameter of flocculating such as zeta potential, cationic demand becomes the bridge ability interrelated with polymeric additive.Because these input-output relations are unknown usually, are that the mathematical control model or the foundation of simulation model all are very difficult (F.Onabe) therefore.F.Onabe, " measuring and control (Measurementand Control) " (the 12nd chapter).W.Scott has explained that these right and wrong are linear, interactional relation.W.Scott, " wet end chemistry principle (Principles of Wet EndChemistry) " (TAPPI publishing house, Atlanta, Georgia, 1996).Wang, H. etc. have described a kind of neutral net, its by wet-end chemicals with the properties relation of generation paper set up model.H.Wang, B.Oyebande, " application (On the Application of Neural Networks Modeling to aWet End Chemical Process in Paper Making) of neural network model in paper making wet part processing " (meeting-proceeding is used in IEEE control, 1995, IEEE, Piscataway, New Jersey, USA, 657-662 page or leaf, 1995).

Realize distributing rationally of control in real time at wet end, can carry out The real time measure is of great use.Usually the mensuration that lacks the timely feedback of relevant adsorption process effect, this is very serious defective, this causes by the control individual parameter input disturbance being reduced to minimum factory's practice.The major part of these parameters, as pH value, ionic demand, concentration (consistency) (Cy), be controlled the flow velocity etc.Yet those performances that are not controlled because lacking online equipment have been brought challenge.For example, ZP generally is a determined off-line.Most of factories lack online ZP analyzer, so can not obtain online ZP control or optimization.

Therefore, a problem relevant with the paper conversion method before the present invention is that they do not provide a kind of and pass through controlled CO ₂Add the method for coming more critically to control wet end chemistry performance in the papermaking in the wet end to.

The another one problem relevant with the paper conversion method before the present invention is that they do not provide controlled CO ₂Add in the wet end, it can control undeterminate character to improve the performance of wet end chemistry in the papermaking.

The again further problem relevant with the paper conversion method before the present invention is that they do not provide CO ₂Add the control mechanism in the wet end to, this mechanism makes more reliable to the control of undetermined character by a kind of improved system disturbance fallout predictor of data that helps is provided.

The present invention seeks to address these problems, and provides a kind of simultaneously and saves cost, paper making wet part easy to use adds CO ₂Control mechanism.

Summary of the invention

The invention discloses a kind of at employing interpolation CO ₂Wet end processing in control CO ₂The method of adding.Paper making compositions and CO ₂In conjunction with, produce a kind of CO that is rich in ₂Paper making compositions.Measure or estimate paper making compositions or be rich in CO ₂At least a electrical property of paper making compositions.Control CO then ₂Interpolation speed make this at least a electrical property remain in the number range of a preliminary election.

In one embodiment, from ZP, CD and ion concentration or any performance suitable, select this electrical property with it, and be selected from flow velocity, CD, ZP, ion concentration, Cy, pH value, electrical conductivity and basicity after measured or paper making compositions or be rich in CO ₂Paper making compositions at least a performance and adopt a kind of model that this electrical property is made estimation.Preferred estimation ZP.Thereby, can or be rich in CO from paper making compositions ₂Paper making compositions, and preferably record the performance of this mensuration from paper making compositions.Similarly, can be paper making compositions or be rich in CO ₂Paper making compositions and be preferably and be rich in CO ₂Paper making compositions estimate the electrical property of this estimation.

By at least a electrical property is remained in the number range of preliminary election, adopt various control schemes so that control CO better ₂Interpolation.

Therefore, an object of the present invention is, by control CO ₂To the interpolation of wet end, provide a kind of method of more critically controlling wet end chemistry performance in the papermaking.

A further object of the present invention is to provide a kind of with CO ₂Add the control mechanism in the wet end to, it can control undeterminate performance, thereby improves the wet end chemistry performance in the papermaking.

Still another purpose of the present invention is to provide a kind of with CO ₂Add the control mechanism in the wet end to, this mechanism by provide a kind of help the improved system disturbance fallout predictor of data make to the control of undetermined performance more reliable, thereby make in the papermaking performance of wet end chemistry better.

The following description of the present invention can make these and other objects of the present invention more obvious.

The accompanying drawing summary

In the following discussion, will be with reference to following accompanying drawing:

Fig. 1 is the sketch that an embodiment of the control method that is suitable for specific paper technology is described;

Fig. 2 is the sketch of another embodiment of this control method of explanation;

Fig. 3 is the sketch of the estimated performance of explanation FEEDFORWARD CONTROL, and it is in response to normalized step-size change (normalized step change) and the disturbance of electric charge demand.

Fig. 4 is the sketch that another embodiment of the control method that is suitable for specific paper technology is described.

Preferred embodiment is described

The invention provides a kind of system and method that is used to control the wet end machined parameters, it injects CO by using first line control unit with real-time mode ₂The design of elder generation's line control unit keeps required setting value, and suppresses the influence of undesirable wet end disturbance.The influence of wet end disturbance can modeling, and adopts their on-line determination so that compensation CO ₂Interpolation.The disclosure provides a kind of effective means for parameter control in the wet end processing, and it uses multivariable control and CO in advance ₂Gas.

As mentioned above, Air Liquide has developed and a kind ofly can utilize CO ₂Change the technology of the zeta potential of cellulose fibre.M.Muguet and J.M.deRigaurd, " improving the manufacturing technique of paper products by the physical and chemical performance that improves paper making raw material ", international publication number No.WO03/074788A2.Air Liquide is in this technical patent application that a pending trial is also arranged, and No. 6052 applications that on September 30th, 2002 submitted as U.S. Provisional Application obtain sequence number 60/414,876, and become U.S.'s non-provisional application on September 6th, 2003, obtain sequence number 10/656,857.The applicant is incorporated herein by reference the full text of these disclosures here clearly, has just done complete elaboration in this application as them.

Of the present inventionly openly relate to a kind of control method that adapts with it.

Since the balance of carbonaceous material, CO ₂Interpolation the basicity in the water is changed.Yet, when at CaCO ₃Exist and add CO down ₂The time, because the generation of extra bicarbonate, basicity changes.

Proved and utilized CO ₂Can improve the efficient of wet end processing.The designing requirement of this application is understood well by CO ₂The interaction that causes and these interactional degree.Relevant knowledge is contained by hydrochemical equilibrium analysis in theory.In order to keep the efficient of this processing, design can keep this processing to move under designed standard and the control system that suppresses main disturbance is helpful.Most important disturbance is attributable to the change of change in concentration, temperature, pressure, concentration or other chemical additive in the compensation water.In order to design the control system that can suppress these changes, know their kinetic effect and they and the CO that supplies to controlled variable ₂Kinetics relation be helpful.Can obtain in these dynamic experiments from obtainable processing knowledge or factory of relation or the theoretical model.

Gaseous state CO is being arranged ₂And CaCO ₃Aqueous system in, can set up the dynamic model of most important material and its influence to zeta potential (ZP) and electric charge demand (CD).The basicity of supposing this dynamic model is constant.

People have studied the CO under the equilibrium condition well ₂-CaCO ₃System, but when being in dynamically (dynamics) condition following time, these researchs just limited to.When exploring the dynamic modeling of this system, need to consider the various parameters that play an important role, as mass transfer and surface reaction.Select CO ₂-CaCO ₃The operating condition of operating has determined how soon reach equilibrium condition, and in design during based on the control system of model, these operating conditions are worthy of consideration.Here various operating conditions have also been listed, such as CO ₂Dividing potential drop, temperature and system's particular variables such as the most important influence of the surface area and the mass tranfer coefficient of calcium carbonate.

The model that checking is advised needs determined off-line basicity and [Ca ²⁺], it is tediously long and consuming time.Therefore in order to predict these determined off-line, considered to use the potential possibility of conductance measurement.

Model development

Balance

At CO ₂-CaCO ₃In the system, the dynamic behavior of material depends on several variablees.This dynamic behavior can use ordinary differential equation (ODE) to express, and it needs primary condition to draw separating of they.Some primary condition can be understood in typical (intermittently) experiment and measure, as the pH value.Yet other primary condition is not understood or is seldom determined.As [CO ₃ ^-2] concentration.Balance studies can provide the estimated value of any component value for any given condition, supposes that these conditions do not change in order to reach balance.Then, EQUILIBRIUM CALCULATION FOR PROCESS can be as the primary condition of ODE.Similarly, EQUILIBRIUM CALCULATION FOR PROCESS can be used for contrasting the final or stable state result of kinetics equation.Under stable state, the result of ODE should be consistent with the equilibrium condition under the new operating condition.

Mass tranfer coefficient, kx, the time response of variation, but do not influence final stable state or equilibrium valve.On the other hand, pCO ₂(CO ₂Dividing potential drop) directly influence the numerical value of balance.Therefore, if possible, need discovery or precision monitor to obtain the pCO of experiment stable state parameter ₂Value.PCO for the operation of finding to reach equilibrium condition ₂, then to solve an equilibrium problem.

At Stumm ﹠amp; Morgan (Stumm, W.and Morgan, " water chemistry: the chemical balance of natural water and speed (Aquatic Chemistry:Chemical Equilibriaand Rates in Natural Waters) " of J.J., John Wiley ﹠amp; Sons, the third edition, 1996) after the 168th page the example, utilize the collection of Matlab program solution by the non-linear algebraic equation of following formula definition:

$\frac{\left[H^{+}\right]\left[\mathrm{HC}{O_3}^{-}\right]}{\left[H2C{O_3}^{*}\right]}=K_1---\left(1\right)$

$\frac{\left[H^{+}\right]\left[C{O_3}^{2-}\right]}{\left[\mathrm{HC}{O_3}^{-}\right]}=K_2---\left(2\right)$

[H ₂CO ₃ ^*]＝K _HpCO ₂ (3)

[H ⁺][OH ^-]＝K _W (4)

[Ca ⁺²][HCO ₃ ^-]＝K _SO (5)

2[Ca]+[H ⁺]＝[HCO ₃ ^-]+2[CO ₃ ^2-]+[OH](6)

Reported the equilibrium constant in the document.

Dependence correlation (the Stumm ﹠amp of the temperature and the equilibrium constant; Morgan)

Log(Ki)＝A1+A2 ^＊T+A3/T+A4Log(T)+A5/T^2

?T(degC)	LogK1	LogK2	LogKH	LogKw	LogKso
						20 25 30 35 40 45 50 55 60 65 70	-6.383133 -6.353105 -6.329367 -6.311358 -6.298577 -6.290573 -6.286946 -6.287334 -6.291415 -6.298898 -6.309522	-10.37557 -10.32885 -10.28788 -10.25227 -10.22169 -10.19583 -10.1744 -10.15714 -10.14381 -10.13418 -10.12806	-1.406868 -1.46794 -1.524416 -1.576641 -1.624926 -1.669556 -1.710789 -1.748859 -1.783981 -1.816351 -1.846147	-14.16818 -13.99953 -13.83949 -13.68751 -13.54308 -13.40575 -13.2751 -13.15074 -13.03233 -12.91953 -12.81207	-8.453297 -8.47983 -8.509751 -8.543034 -8.579652 -8.619574 -8.662766 -8.709195 -8.758826 -8.81162 -8.867541

The dependence of the equilibrium constant and temperature:

Stumm ﹠amp; Morgan has reported the correlation that temperature relies on.

Equation (1)-(6) have been described at given CO ₂Dividing potential drop (pCO ₂) under, the CO in the water ₂-CaCO ₃Equilibrium equation.These equations can be different form use, to adapt to different conditions.For example,, can not use equation (5) so, need to estimate [Ca from rate equation if calcium is not in poised state ^{+ 2}].

Dynamics (speed)

Balance studies is determined limit or based on total variation of operating condition.And on the other hand, dynamics research also determines to take place the speed of this change.This two kinds of results, the degree of variation and the speed of variation are depended in the process control design.Under the condition of constant basicity, CO ₂The diffusion equation that (bulk) liquid exchange can be expressed as from the gas phase to the body:

$\frac{d{C}_T}{\mathrm{dt}}=\mathrm{klx}(p_{C{O}_2}H+\mathrm{Alk}-C_T)=R_{{\mathrm{CO}}_2}---\left(7\right)$

Wherein,

C _T=whole carbonaceous materials, M=[H ₂CO _3*]+[HCO ₃ ^-]+[CO ₃ ^-2]

The klx=mass tranfer coefficient, minute ^-1

PCO ₂=CO ₂Dividing potential drop in gas, atm

The H=Henry equilibrium constant

Alk=basicity, M (mole HCO ₃/ liter)

Equation (7) is at [CO _3--] concentration can ignore and C _TBe effective during＞＞Alk.As pressure CO in the atmosphere ₂Dividing potential drop the time, will this thing happens.When pressure during near atmospheric pressure, equation (7) is invalid, but supposition can adopt the operating condition of this equation for being rich in CO ₂Condition.

If basicity is not constant, CO so ₂Be converted into [H ₂CO _3*] the mole transfer rate as shown in the formula:

$\frac{d\left[{\mathrm{<figure-callout\; id="2"\; label="H"\; filenames="C20048001688600381.png"\; state="\{\{state\}\}">H</figure-callout>}}_2{{\mathrm{CO}}_3}^{*}\right]}{\mathrm{dt}}=\frac{d}{\mathrm{dt}}[C_T-[{{\mathrm{HCO}}_3}^{-}\left]\right]=\frac{d{C}_T}{\mathrm{dt}}-\frac{d\left[{{\mathrm{HCO}}_3}^{-}\right]}{\mathrm{dt}}---\left(8\right)$

Then, all the speed of carbonaceous material becomes:

$\frac{d{C}_T}{\mathrm{dt}}={R_{\mathrm{CO}}}_2-\frac{d\left[{{\mathrm{HCO}}_3}^{-}\right]}{\mathrm{dt}}={R_{\mathrm{CO}}}_2-\frac{\mathrm{dAlk}}{\mathrm{dt}}---\left(9\right)$

In order to estimate the pH value over time, can from following relational expression, begin:

$\left[H^{+}\right]=\left(\frac{\mathrm{<figure-callout\; id="1"\; label="p\; K"\; filenames="C20048001688600371.png,C20048001688600391.png"\; state="\{\{state\}\}">p</figure-callout>}{K}_{}{}_1}{\mathrm{Alk}}\right)(C_T-\mathrm{Alk})---\left(10\right)$

Can be rewritten into:

$\left[H^{+}\right]=p{K}_1\frac{C_T}{\mathrm{Alk}}-p{K}_1---\left(11\right)$

Equation (11) is carried out the time differentiate shows that when considering the time dependence of basicity, this equation becomes more complicated:

$\frac{d\left[H^{+}\right]}{\mathrm{dt}}=p{K}_1\frac{d}{\mathrm{dt}}\left(\frac{C_T}{\mathrm{Alk}}\right)---\left(12\right)$

It becomes:

$\frac{d\left[H^{+}\right]}{\mathrm{dt}}=p{K}_1\left\{\frac{\mathrm{Alk}*R_{{\mathrm{CO}}_2}-C_T\mathrm{dALK}}{{\mathrm{Alk}}^2}\right\}---\left(13\right)$

The calculating of equation (13) depends on equation (9) or R _CO2, but also depend on dAlk/dt.At present, also do not find the analysis expression of dAlk/dt.

When at CaCO ₃Exist and in system, add CO down ₂The time, the proton equilibrium expression formula of available simplification draws a tentative expression formula:

2[Ca ⁺⁺]～＝Alk (14)

Though equation (14) is a kind of equilibrium equation of proton equilibrium, can be all the time represent approximation relation between calcium ion and the basicity with it.Therefore, the equation both sides are to the time differentiate,

$\frac{\mathrm{dAlk}}{\mathrm{dt}}=2\frac{d\left[{\mathrm{Ca}}^{++}\right]}{\mathrm{dt}}---\left(15\right)$

Because numerical computations uses equation (13) and (15) and cause application problem, and the result is unsuccessful.When solving an equation (13) and (15) simultaneously, the concentration of hydrogen proton becomes non-real number.Therefore, a kind of simplification appears.

Known proton transfer reaction, very fast usually as the proton transfer reaction (carbonic reaction) of carbon containing, its half-life is less than several milliseconds.These prompting people can represent in ODE that this equation is to estimate hydrogen proton or basicity.Instead, when the speed defining equation among the ODE solves, can determine to find the solution the residual chemicals of this appropriate balance equation.

C has correspondingly been described in equation (9), (13) and (15) _T, pH and Alk kinetics equation.The calcium dissolution kinetics that people such as Plummer obtain has following form:

R _Ca＝K _Ca，1α _H+K _Ca，2α _H2CO3+K _Ca，3α _H2O-K _Ca，4α _Caα _HCO3(16)

Plummer, L.N., Wigley, T.M., and Parburst, " calcite is dissolved in 5-60 ℃ and 0.0-1.0atm CO to D.L. ₂CO ₂Dynamics in the-water system (Thekinetics of calcite dissolution in CO ₂5 ° of to60 ℃ of and of-water systems at, 0.0 to, 1.0 ATM CO ₂) " (American Journal of Science, the 278th volume, 179-216 page or leaf, in February, 1978).Activity coefficient, a _i, equal concentration.In the time of 25 ℃, parameter is:

k _Ca，1＝5.115e-02；

k _Ca，2＝3.4247e-05；

k _Ca，3＝1.1919e-07；

k _Ca，4＝4.55e-02；

R _CaUnit be mmol/ (sec ^-1Cm ²).Constant k _{Ca, i}Depend on temperature, and k _{Ca, 4}It is backward reaction.Except transforming R _CaFor mole and minute ^-1Outside, this dynamics expression formula depends on the total surface area of calcite particle.With the particle cartographic represenation of area is a _c, be defined as the surface area of per unit mass, as the calcium dissolving represented with moles/min as shown in the formula:

$\frac{d\left[{\mathrm{Ca}}^{+2}\right]}{\mathrm{dt}}=a_c\left[\mathrm{Ca}{\mathrm{CO}}_3\right]V{R}_{\mathrm{Ca}}(60/1000)---\left(17\right)$

Product [CaCO ₃] V is equivalent to CaCO in reactor ₃Total gram number.Plummer has reported that the scope of the calcite particle that polishes is up to 90 centimetres ²/ gram.

These CO ₂It is the phenomenon of expression overall rate that the speed of migration and calcium dissolving is expressed.Therefore, in order to discern CO ₂-CaCO ₃The time dependence value of all substances in the system, little increment is found the solution CO ₂The ODE (equation (9) and (17)) of migration and calcium dissolving can return [Ca ^{+ 2}] and whole carbonizable substance, C _TThen, find the solution following algebraic equation and can draw basicity, [HCO _3-] and hydrogen proton [H ⁺]:

$\frac{\left[H^{+}\right]\left[{{\mathrm{HCO}}_3}^{-}\right]}{\left[{\mathrm{<figure-callout\; id="2"\; label="H"\; filenames="C20048001688600381.png"\; state="\{\{state\}\}">H</figure-callout>}}_2{{\mathrm{CO}}_3}^{*}\right]}=K_1---\left(18\right)$

$\frac{\left[H^{+}\right]\left[{{\mathrm{CO}}_3}^{2-}\right]}{\left[{{\mathrm{HCO}}_3}^{-}\right]}=K_2---\left(19\right)$

[H ⁺][OH ^-]＝K _W (20)

C _T＝[H ₂CO ₃ ^*]+[HCO ₃ ^-]+[CO ₃ ^2-] (21)

2[Ca]+[H ⁺]＝[HCO ₃ ^-]+2[CO ₃ ^2-]+[OH ^-] (22)

In case solve these equations, available new numerical solution goes out the next increment of ODE.

Then, CO ₂-CaCO ₃Dynamic model can define by 2 ODE (equation (9) and (17)) and simultaneous solution algebraic equation (equation (18) is to (22)).All these equations have had the various parameters of essential appointment before numerical simulation.These parameters are temperature, mass tranfer coefficient, CO ₂Dividing potential drop and the surface area of calcium carbonate.

The all kinetic constant and the equilibrium constants have dependence to temperature.The temperature dependency of these kinetic parameters has been proposed in the former report (CRC200343), and with T=25 ℃ list as above.Listed the temperature dependency of balance parameters in the appendix A.

Experiment

Carry out a series of experiments.8.125 gram CaCO in the deionized water that joins 1.3 liters are adopted in these experiments ₃(being equivalent to the 20%PCC in the paper pulp).Use two types CaCO ₃: ALBACAR HO PCC (Specialty Minerals, Inc) and the CaCO of SILVER REAGENT ₃(FisherScientific).Stir CaCO ₃Storing solution 24 hours with reach with atmosphere in CO ₂Balance.From this storing solution, get first sample (time is 0).1.3 liters storing solution is put into reactor, and mixing speed is set in 1500rpm.Work as CO ₂When the metering feed begins, pick up counting.Be assumed to 2.5%Cy slurry or 2.6e ^-04Kilogram CO ₂/ when rising, total CO ₂Amount is 10.256 kg/ton, to finish CO in about 40 seconds ₂The metering feed.The purpose of measuring is will obtain at the constant CO of reactor head ₂Dividing potential drop.Taking-up is near 10 milliliters sample, at once with 0.02 micron filter filtration.In order to obtain more sample, for several times with identical condition experiment.Measure electrical conductivity, pH value, basicity and the calcium concentration of each sample.

Experimental result

Table 1 has shown the experimental result of use ALBACAR PCC～24 ℃ of acquisitions.Be apparent that especially after a few minutes, the concentration and the basicity of calcium increase fast, reach equilibrium valve.

Table 1 uses the experimental result of ALBACAR PCC (24 ℃)

The proton equilibrium that shows by equation (6) and since under this pH value condition the concentration [CO of carbonate _3--] can ignore, below concern: 2*[Ca ^{+ 2}The Alk of]～=(Stumm ﹠amp; Morgan) necessarily be satisfied.As can be seen, this relation is not met from the experimental result that table 1 shows.Compare with the equilibrium phase of calcium dissolving, these results represent much higher basicity value.These results show that ALBACAR PCC has some influences CO ₂-CaCO ₃The chemicals of balance.

Table 2 has shown use SILVER REAGENT CaCO ₃Experimental result.As ALBACAR, the concentration of calcium and basicity increase very soon towards the balance direction.The back will provide the more deep explanation to this behavior.About balance, the interesting fact is that the relation between balance calcium and basicity more approaches to satisfy relational expression: 2*[Ca ^{+ 2}The Alk of]～=.Except 1 minute sample, the mean error between experiment value and the analysis expression is 5%.The higher experimental error of 1 minute sample is had any problem owing to obtaining representative sample at such experiment initial stage.

Table 2 uses CaCO ₃Experimental result (25 ℃)

It should be noted that in the time to be that the pH value of filtered sample changes very little in experimentation between 0 sample and the last sample.According to equilibrium analysis, the pH pH-value determination pH value of sample is more near desired value before filtering.

The result that table 2 is listed is used for the remainder of this research, and balance and dynamic model are compared.

Result's equilibrium analysis

Equilibrium equation can be determined CO ₂-CaCO ₃The target limit of system.Use these information, people can be in experiment and the dynamic model simulation in find the stable state numerical value of expectation in advance.

The first step, confirm ask a question and result's validity, it is the repetition of Stumm and Morgan problem on 186 pages.This problem is to be determined at atmospheric pressure (10 ^-3.5Atm=3.16e-04atm) CO under ₂-CaCO ₃Equilibrium concentration in the system.(Mathworks, Inc) the non-linear algebraic equation solving program command of middle use can be finished finding the solution of this problem at Matlab.Solving equation collection (1)～(6) among the Matlab and the result who draws is:

[Ca]＝4.636e-04M；

Alk=9.2721e-04M and

PH value=8.3048.Should also be noted that 2[Ca ²⁺The Alk of]～=.

In order to use CaCO ₃The experimental result that obtains is compared with equilibrium equation, needs CO ₂Dividing potential drop.This pressure is unreliable always experimentally maybe can not to be obtained.Some early stage mensuration is pointed out CO ₂Branch is about 0.43psig when being pressed in the experiment beginning, equals 0.02atm, but its continuous decline and its record are unreliable.In addition, as equation (3) representative, this dividing potential drop feature of system is assumed to be constant during balance.This is not the situation of system under the experimental provision condition, and promptly system is a batch reactor, and the variable time to time change.Yet in order to obtain equilibrium equation, people have to seek the feature CO that reaches the identical experiment equilibrium condition ₂Dividing potential drop.Table 3 has shown for certain CO ₂The dividing potential drop scope, the one group of result who obtains during solving equation (1)-(6).

Table 3CO ₂-CaCO ₃The equilibrium condition of system (25 ℃).

Will be from comparing with the experimental result that table 2 obtains that equilibrium equation obtains as the listed balance result of table 3 (using extrapolation), draw the equilibrium condition that meets experimental result and be:

Situation A: 25 ℃ of theoretical balances.

I.Alk=5e-03 mole HCO ₃/ liter=250 milligrams of CaCO ₃/ liter=250ppmCaCO ₃

Ii.[Ca]=2.5e-03 mole Ca/ liter=100 milligrams of Ca/ liters

iii.pCO ₂＝10 ^-1.049atm＝～0.09atm

Iv.pH value=6.5 (Matlab), 6.9-7.0PHREEQ

(Specialty Minerals, experimental result Inc.) shows that what indicate balance is the pH value to use ALBACAR HO.But obtain the CO of 7.3 balance pH values ₂Dividing potential drop has reached the theoretical basicity lower than experiment value.Theoretical equilibrium condition in this case is designated case B, and as follows.About this point, people's conjecture, result's difference may be because used PCC (ALBACARHO, Specialty Minerals, Inc) middle unknown component.

Case B: final balance test result (12/12/03)

I.pH value=7.3

ii.[Ca]＝106ppm＝2.7e-03M

Iii.Alk～=300ppm is as CaCO ₃=6e-03 mole [HCO ₃]/liter

Iv. initial pCO ₂～0.4psig (+/-0.2psig)=2.72e-02atm

Even balance pH value theoretical and under the experiment situation is the same, but departing from of calcium concentration and basicity is very remarkable.Calcium concentration experimentally almost is 2 times of theoretical value.Theoretic balance calcium concentration depends on constant K _SoExcept experimental error, a kind of possibility of deviation is used CaCO ₃Perhaps impure, perhaps different.Based on experimental result, and, can obtain some former equations merging:

$\left[C{O_3}^{-2}\right]=\frac{K_1{K}_2{K}_H\mathrm{pC}{O}_2}{{\left[H^{+}\right]}^2}=2.8637e-06M---\left(23\right)$

In the substitution equation (5), use experiment calcium concentration (2.65e-03M), can obtain equilibrium constant K _So:

K _so＝[Ca ⁺²][HCO ₃ ^-]＝7.588e-09＝10 ^-8.1198(24)

This draws a K _So, it almost is 2 times of bibliographical information.In the time of 21 ℃, with new K _SoWith other constant operation balance software, can the results are as follows:

Situation C: adopt new K _SoTheoretical value

I.pH value=7.3061

ii.pCO ₂＝10 ^-1.985atm＝1.0351e-02atm

iii.[Ca]＝2.2e-03M＝89.75ppm

iv.Alk＝4.5e-03M＝274.5ppm

Although the balance calcium concentration of estimating is now more near experiment value, the basicity estimated value is subjected to negative effect.Following table has been listed gathering of these results.

Situation	T (℃)	pCO 2 (atm)	The pH value	[Ca](M)	Alk(M)	Remarks
							text	25	10 -3.5	?8.304	4.636e-04	9.2721e-04	As shown in the Morgan
text	25	10 -3.5	?8.304	4.636e-04	9.2721e-04	In Matlab, repeat
							A	25	10 -1.04	?6.8	2.5e-03	5e-03	In theory
B	21.3	0-10 -1.45	?7.3	2.7e-03	6e-03	ALBACAR PCC experiment
							C	21.3	10 -1.985	?7.306	2.2e-03	4.5e-03	New K so
D	21.3	10 -2.05	?7.3	1.4e-03	2.8e-03	In theory
							E	25	N/A	?7.1(？)	2.46e-03	5.08e-03	Chemical grade CaCO 3Experiment

This table demonstrates and uses chemical grade CaCO ₃Experimental result as one man satisfy the relation (situation E) of calcium and basicity.Here, will use the dynamic experiment data that from this experiment, get to compare with dynamic model.Will use the complete operating condition group that comes from EQUILIBRIUM CALCULATION FOR PROCESS (situation A), as pCO ₂

Result's dynamics (speed) is analyzed

Mass tranfer coefficient kx in equation (9) has defined gas CO ₂Move to the speed in the liquid.It depends on the design of physical property and equipment.For specific design and operating condition, can find some correlations in the literature.Based on disclosed document, near the stirred tank of gas bubbling agitator, typical mass tranfer coefficient is approximately 0.2 minute therein ^-1In this below simulation of value as reference.

It is found that when dividing potential drop was 0.09atm, the listed experimental result of table 2 was exactly an equilibrium condition.This pressure can be used as nominal operation pressure.

At last, need surface area to determine rate of dissolution according to equation (16).When discovery was～1.6 microns in average diameter size, (Specialty Minerals, surface area Inc) equaled 11.5 meters to ALBACAR OH ²(11.5e+04 centimetre of/gram ²/ gram).On the other hand, derive from the CaCO of Fisher ₃Average diameter be 30～50 microns.This may represent surface area 10 ⁴Centimetre ²The scope of/gram.Maximum surface area that it should be noted that the calcium carbonate of reporting in people's such as Plummer calcium dissolving paper is 90 centimetres ²/ gram, this with CRC experiment in the calcium carbonate measured compared several orders of magnitude little.To introduce the conclusion of surface area and integral power in the calcium dissolving after a while.

The dynamic model experiment

Use the nominal operation condition:

i.pCO ₂＝0.09atm

Ii.kx=0.2 minute ^-1

iii.T＝25℃

Iv.a _cThe surface area of=calcium carbonate=10 ³Centimetre ²/ gram

This set condition will be designated condition group A.

Primary condition is corresponding to adopting Atmospheric CO ₂Dividing potential drop (10 ^-3.5Atm) equilibrium condition:

PH value=8.96; Alk=3.3e-04 mole HCO ₃/ liter; [Ca ^{+ 2}]=1.65e-04 mole Ca/ liter; C _T=3.45e-04 mol.In this report, all basicity can be expressed as mol, molar average (meaning mol) [HCO _3-]/liter, and calcium can be expressed as mol, molar average Ca/ liter.

Table 4 has shown the simulation model of the service condition group A that contrasts with empirical value.Work as CO ₂The nominal dividing potential drop when 0.09atm is reduced to 0.07atm, prediction of the stable state of model and experimental data coupling.When in order with equilibrium equation it to be found the solution dynamic model being dispersed, this difference is owing to the loss of precision.And, because require " removing error (troubleshooting) ", thereby be difficult to draw separating that table 4 lists.When finding the solution ODE (equation (9) and (17)), it should be noted that separating of dCa/dt can cause [Ca ²⁺] value is higher than the numerical value that can obtain in balance.The dynamic resolution of dCa/dt can not solve the concentration higher than balance in native system.Under limit, dCa/dt should only equal the prediction of balance.The conclusion that this problem produces is: when the calcium carbonate surface area apparently higher than (90 centimetres of the used surface areas of people such as Plummer ²/ gram) time, the dissolving that the dynamics expression formula of equation (16) has been over-evaluated calcium.

Surface area height (～10 when calcium carbonate ⁴Centimetre ²/ gram or higher) time, the dissolving of calcium is very fast, so that its in fact instantaneous balance that reaches, this makes this reaction the same fast with the proton transfer reaction.This quick calcium dissolving prompting people calculate the calcium of any preset time with equation (5) equilibrium equation of being given, and adopt other proton transfer reaction simultaneously.Like this, for high surface, at each time interval (dC _T/ dt), only there is a differential equation to be found the solution.In case know C by equation (9) a sampling interval _T, will solve following equation and find out the balance calcium concentration:

${\mathrm{<figure-callout\; id="1"\; label="K"\; filenames="C20048001688600371.png,C20048001688600391.png"\; state="\{\{state\}\}">K</figure-callout>}}_1=\frac{\left[H^{+}\right]\left[\mathrm{HC}{O_3}^{-}\right]}{\left[H_{2}C{O_3}^{*}\right]}---\left(25\right)$

${\mathrm{<figure-callout\; id="1"\; label="K"\; filenames="C20048001688600371.png,C20048001688600391.png"\; state="\{\{state\}\}">K</figure-callout>}}_1=\frac{\left[H^{+}\right]\left[\mathrm{HC}{O_3}^{-}\right]}{\left[H_{2}C{O_3}^{*}\right]}---\left(26\right)$

K _S＝[Ca ²⁺][CO ₃ ^2-] (27)

C _T＝[H ₂CO ₃ ^*]+[HCO ₃ ^-]+[CO ₃ ^2-]?(28)

These equations can merge to be obtained only as C _T[the Ca of function ²⁺]:

$C_T=\frac{4K_2}{K_1{K}_S}{\left[{\mathrm{<figure-callout\; id="2"\; label="Ca"\; filenames="C20048001688600381.png"\; state="\{\{state\}\}">Ca</figure-callout>}}^{2+}\right]}^3+2\left[C{a}^{2+}\right]+\frac{K_S}{\left[C{a}^{2+}\right]}---\left(29\right)$

Equation (29) is not owing to need the initial value iteration, so find the solution sooner, this is as finding the solution the situation of simultaneous non-linear algebraic equation.Analog result in the table 4 shows the calcium of being calculated by equilbristat, because the dynamics prediction is higher than the value of balance, is impossible so believe this.

CO ₂Dividing potential drop and temperature effect rate of dissolution, and influence equilibrium condition.In this, people are only interested in change speed, and keep balance (stable state) result according to experimental result.The unique obtainable parameter that can reach this purpose is a mass tranfer coefficient.Experiment is carried out under very high-speed (1500rpm), but mass tranfer coefficient is not known.If mass tranfer coefficient was from 0.2 minute ^-1Be increased to 2.0 minutes ^-1, CO then ₂Move to body quickly, and the proton transfer reaction takes place more early.

Table 5 shows the analog result that obtains with higher mass tranfer coefficient.Can see that now this experimental result is more consistent with analog result.

Conclusion is to have developed CO ₂-CaCO ₃The dynamic model of system is used to the experimental result of representing CRC to obtain.Yet, must emphasize, for good analog result, must consider correct operating condition knowledge, as: the characteristic of dividing potential drop and mass tranfer coefficient and calcium carbonate, so that obtain the best model representativeness.Experimental example makes this more modeled key characters obtain explanation.Next, being explained in more detail of these modeling problems will be listed.

Table 4 simulation model-experiment (condition group A)

Table 5 mass tranfer coefficient, kx=0.4 minute ^-1The time simulation model

Surface area

The dissolution kinetics of calcium is subjected to the control of surface reaction, and this makes surface area have importance.Demonstrate, for the calcium carbonate of high surface, a kind of good approximate be the prediction of ignoring calcium speed, and instead with equilibrium equation.However, still propose under the situation of little surface area, about the more deep opinion of calcium dissolving.From now on, will list several simulations, defined condition among the following condition group B is used in these simulations:

Condition group B:

i.pCO ₂＝0.025atm

Ii.kx=0.2 minute ^-1

iii.T＝25℃

Iv.a _c=le+05 centimetre ²/ gram

Table 4 and table 5 have shown some simulations, the surface area high significantly (10 of these simulation hypothesis calcium ⁴Centimetre ²/ gram+), this makes dissolving make progress so soon, so that almost reach balance at once, and calculate calcium from equilibrium equation.Table 6 has shown when surface area some analog results hour, has then described this calcium dissolving with the kinetics equation of people such as Plummer report.

Table 6CaCO ₃The influence of surface area (ac with centimetre ²/ gram meter)

Usually, can see (90 centimetres of minimum surface areas ²/ gram) can need the time of a long number order of magnitude to reach balance.The highest face temperature of simulation is amassed 9e+04 centimetre ²In the equilibrium equation and kinetics equation of/gram substitution combination.In the simulation beginning with till about 5 minutes, the prediction of speed value is higher than equilibrium valve, and this is impossible, has therefore reported the balance forecast value.After 5 minutes, the prediction of speed value is lower than equilibrium valve, therefore, and to the prediction of speed value mapping of Plummer etc.Conversion between a model and another model causes some digital deflections, and it is shown as spike in these simulations.

Note the influence of surface area to the pH value.The surface area height can strengthen the calcium dissolving and make the proton transfer equation to lower [H ⁺] concentration (high pH value) direction balance.On the other hand, when surface area hour, the calcium dissolving is very slow, and the equilibrium concentration of proton transfer material, as [H ⁺], similar when almost not having calcium.For this reason, table 6 has shown the CaCO that ought in fact not have calcium and will not dissolve ₃The time simulation.When there not being CaCO ₃The time, it is constant that basicity and calcium keep, and the pH value drops to the value of lower stable state fast.

Table 7 has shown the view of the table 6-pH value that distance is nearer.Can be more clearly visible, when at the surface area of initial time hour, this system puts up a good show to resemble does not have calcium dissolving and the decline of pH value very fast.When calcium concentration begins to increase, pH value beginning towards with CaCO ₃Corresponding balance direction increases.

Table 7CaCO ₃Surface area to the influence of pH value (ac with centimetre ²/ gram meter)

Mass tranfer coefficient

Mass tranfer coefficient can not influence stable state or equilibrium condition.As the calcium carbonate surface area, it only influences and will how soon just can reach balance.Consider the time constant of these variable influence processes and do not influence gain from the viewpoint of process control.Table 8 has shown the influence of different mass tranfer coefficients on all variablees.

The influence of table 8 mass tranfer coefficient kx

Temperature

Influence a temperature that factor is a system of equilibrium condition.The all equilibrium constant and kinetic constants all have dependence to temperature, and any variation all can be influential to concentration.Table 9 has shown a pair of simulation under different temperatures.Shown maximum temperature (50 ℃) drives reaction and moves towards lower calcium dissolving direction.Equilibrium constant value in the appendix shows, the temperature effect constant K ₁, K ₂And K _SMove to different directions.

About the sensitiveness of variable for temperature, table 9 shows, temperature to the influence of concentration greater than influence to the pH value.But little pH value changes representative [H ⁺] big change (because logarithmic relationship) on the change in concentration.If in factory little variations in temperature takes place, when using CO ₂The time, do not expect big performance change.

Table 9 Temperature Influence

CO ₂ Dividing potential drop

Table 10 shows C0 ₂Dividing potential drop is to CO ₂-CaCO ₃The influence of system.The dividing potential drop increase has promoted CO ₂Migration, the reduction of pH value and the dissolving of calcium.Dividing potential drop also directly influences limit (balance).Different with temperature, CO ₂The less relatively variation meeting of dividing potential drop comprises that to all variablees the pH value produces bigger influence.It continues to play a part most important manipulated variable, is stipulating final operating condition.

Table 10pCO ₂Influence

Electrical conductivity

(Specialty Minerals in experimentation Inc), writes down online conductance measurement value using ALBACAR OH.Table 11 has shown use ALBACAR PCC (SpecialtyMinerals, CO Inc) ₂-CaCO ₃The online conductance measurement value of one of experiment.Electrical conductivity is the tolerance that the aqueous solution transmits current capacity.This ability depends on the existence of ion; The total concentration, mobility and the chemical valence that depend on them; And depend on the temperature of mensuration.People such as Clesceri have proposed to calculate the general theory method of electrical conductivity.Referring to Clesceri, L.S., Greenberg, A.E., and Eaton, " standard test method of water and waste water (the Standard Methodsfor the Examination of Water and Wastewater) " of A.D. (the 20th edition, APHA, AWWA, WEF).At first, calculate the electrical conductivity of infinite dilution liquid:

$k^o=\mathrm{\&Sigma;}\left|z_i\right|\left({\mathrm{\&lambda;}}_{+I}^0\right)\left(m{M}_i\right)+\mathrm{\&Sigma;}\left|z_i\right|\left({\mathrm{\&lambda;}}_{-i}^0\right)\left(m{M}_i\right)---\left(30\right)$

Wherein:

| z _i| the absolute value of=the i ionic charge

MM _iThe millimolar concentration of=the i ion

Then, calculate ionic strength, IS is in molal unit:

$\mathrm{IS}=\mathrm{\&Sigma;}{z}_i^2\left(m{M}_i\right)/2000---\left(31\right)$

The activity coefficient of monovalention, y calculates with the Davies equation, and it is applicable to that IS＜=0.5M and temperature are 20 ℃ to 30 ℃.

y＝10 ^{-0.5(a-0.3IS)}； $a=\frac{{\mathrm{IS}}^{0.5}}{1+{\mathrm{IS}}^{0.5}}---\left(32\right)$

At last, electrical conductivity can be calculated with following formula:

K _calc＝K ⁰y ² (33)

Can obtain the equivalent ionic conductivity the aqueous solution from different sources, comprise " Lange ' s Handbook of Chemistry ".Dean, J.A. " Lange ' s Handbookof Chemistry " (McGraw-Hill, the 13rd edition).Above-mentioned equation is that the function of electrical conductivity as the effects of ion molar concentration calculated, and uses above-mentioned equation to come " inverse " ion concentration based on conductivity measurements then.Owing in experiment, use deionized water, so in solution, only can find two kinds of main components as can be known: [Ca ²⁺] and [HCO ₃ ^-].Yet, have the ion concentration combination that can produce the same electrical conductance of unlimited amount.For this reason, need additional relational expression.In the pH value scope of being concerned about, CO ₂-CaCO ₃Proton equilibrium equation in the system can be reduced to:

2[Ca ²⁺]～＝[HCO ₃ ^-]＝Alk (34)

So, adopt and calculate [Ca ^{+ 2}] and [HCO ₃ ^-] ion concentration optimization problem and be defined as follows:

$\underset{\left[C{a}^{2+}\right],\left[\mathrm{HC}{O_3}^{-}\right]}{\min }J=\min {(k_{\exp }-{\hat{k}}_{\mathrm{calc}})}^2---\left(35\right)$

It makes equation (32) for true.By minimizing the objective factor (objective factor), J, this inverse is found the solution.This objective factor is by electrical conductivity K that measure or experiment _ExpWith the electrical conductivity (K that calculates by equation _Calc) the mean square error of estimated value form.This separates the minimizing of this problem of satisfying the proton equilibrium equation with being.In Matlab, this problem is carried out numerical solution with minimizing of a multivariable constraint function.

Table 12 be presented at one use ALBACAR PCC (Specialty Minerals, in experiment Inc), the online conductance measurement value of experiment and the [Ca of off-line ²⁺] and the basicity measured value.Except first off-line sample since sampling difficulty and may be considered to inaccurate, the predicted value of the calcium that is obtained by the conductance measurement value and the measured value of experiment of off-line very closely mate.On the other hand, the evaluated error of basicity is bigger.Worst error can reach 40 units.

The online electrical conductivity that table 11 adopts the experiment of ALBACAR OH PCC to record

Table 12 uses electrical conductivity and the predicted value of ALBACAR OH

Table 13 has shown and has wherein used chemical grade CaCO ₃The basicity of experiment and [Ca ²⁺] estimated value.The estimated value of calcium is almost the same with last situation good.On the other hand, the predicated error of basicity is less.This means that this experiment is made up of the ion near desired value more.Use the experiment of ALBACAR to have some composition that influences electrical conductivity or ions, and may repel equation (32).In a word, we can say that the mensuration of electrical conductivity has prediction CO ₂-CaCO ₃The potentiality of most important ion development in the system.

Table 13 is from chemical grade CaCO ₃Electrical conductivity and predicted value

In the preferred embodiment of describing like this, help to determine CO ₂-CaCO ₃The phenomenon of each material speed is CO in the system ₂Migration and the dissolving of calcium.Other reaction is to take place fast also can think that they are in the proton transfer reaction of poised state, and it comprises basicity.Present disclosure has proposed the calculating of proton transfer reaction, and its each integration step at rate equation is calculated and proofreaied and correct.

(surpass 10 if the surface area of calcium carbonate is big significantly ⁴Centimetre ²/ gram), the dissolving that then can calculate calcium with equilibrium equation.Used CaCO in the experiment of this report ₃Have preferably predicting the outcome of high surface area and the EQUILIBRIUM CALCULATION FOR PROCESS of considering calcium.Along with CO ₂The basicity brought of supply increase the pH value had big influence, thereby the dissolving of calcium is also had big influence.

Once calculated basicity, to CO in time dependent mode ₂Dividing potential drop, temperature, mass tranfer coefficient and calcium carbonate surface area are in the CO of various conditions ₂-CaCO ₃System is simulated.If the calcium carbonate surface area is big, then controlling the most important factor of this system's speed is mass tranfer coefficient.If this surface area little (particle diameter is big), then because the calcium rate of dissolution is slower, the time constant of this system can remarkably increase.

Different with mass tranfer coefficient and granule surface area, the temperature of system and CO ₂Dividing potential drop influence balance (stable state) condition.Dividing potential drop is bigger than temperature to the influence of equilibrium condition, therefore possesses the good CO of control ₂Operation be important.

Show also that at last the mensuration of electrical conductivity can be used for predicting CO ₂-CaCO ₃The calcium of system and basicity.If have other ionic species to exist, then can need more detailed derivation.

With reference now to Fig. 1,, what it was shown is the sketch that an embodiment of the control method that is suitable for specific paper technology is described.The CO that blending chest 1 receives through regulating ₂Inlet stream and paper making compositions, such as fibre stream, 7, and provide and be rich in CO ₂Paper making compositions, such as the outlet fibre stream.The performance of fibre stream, for example flow velocity, CD (electric charge demand), ZP (zeta potential), ion concentration, Cy (concentration), pH value, electrical conductivity and basicity online measurement before this fibre stream arrives this blending chest, and produce measured value 2.Simultaneously, measure the on-line performance that this exports fibre stream with instrument 3, and these on-line performances can be selected from one group of performance same as described above.Elder generation's line control unit is a feedforward controller under this situation, uses the measured value of inlet stream, and by adding controller output valve, 6, compensation feedback control device, 5.The controller output valve that produces is used for handling inlet CO ₂, it will keep required wet end performance, and the change of inlet fibre stream is minimized.

It should be noted that some performance can be described as electrical property, for example ZP, CD and ion concentration.Because ion is atom or the molecule that has obtained or lost one or more electronics, so it has clean negative electrical charge or positive charge.For example, fusionplasma is so warm, so that in fact all electronics are peeled off from atom, thereby produces a kind of ion, and the clean positive charge of this ion equals the proton number in its atomic nucleus.The quantity of this ion in ion concentration and the per unit volume is relevant.Therefore, there is the direct correlation of the concentration of charged ion.

Yet if be mixed with two kinds of different ion A, B in solution, its concentration is [A] and [B], still has a kind of total electrical charge, but for example measures this total electrical charge with conductivity meter in solution, can not distinguish and measure [A] and [B].Present disclosure is expected respectively and is distinguished and measure [A] and [B] individually.What should be noted usually is, ion concentration is as used herein, and this operation is carried out in expection when needed.

Fig. 2 has shown the block diagram of feedforward controller.It needs disturbance and wet end to measure the mathematical relationship between (in this case, being zeta potential (ZP), GL) and the CO that enters the mouth ₂With the relation between the identical wet end mensuration Gp.The design of elder generation's line control unit is to find the mathematical relationship between feedforward controller F and the feedback controller Gc.

The present embodiment can be applicable to the wet end of pulp mill, controls the character of paper by the controlled interpolation of additive.More specifically, the present embodiment has been considered CO in the blending chest ₂Interpolation at wet end.The measurement of most important variable such as zeta potential, electric charge demand, basicity, pH value etc. carries out, collects and makes data collecting system can obtain these measured values in real time in the entrance and exit fibre stream.Adopt the mathematical relationship of transfer function form or any time forms of dependency to make CO ₂Interpolation and desired wet end performance or controlled variable, be associated as zeta potential or electric charge demand.In addition, also adopt similarly relation between possible variation or disturbance and the controlled variable at inlet in the fibre stream.Employing tends to make the required controlled value and the minimized any feedback controller of the deviation between the The real time measure value of outlet fibre stream.

In addition, will be used to design predictive controller from the obtainable The real time measure value of inlet fibre stream, such as feedforward controller, model predictive controller (MPC) or any first line control unit.Feedforward controller is revised the output valve of feedback controller with cooperative mode.Feedback and feed-forward signal stack, and be sent to linear proportional control valve, this valve control enters the CO of mixing channel ₂Stream.

Fig. 3 shows the performance in response to the FEEDFORWARD CONTROL of normalization step-size change and the disturbance of electric charge demand.Can see that feedback controller is no problem for the response that setting value changes, and because feedforward controller, this disturbance almost is suppressed, and makes device turn back to the setting value of hope at once.

Be used to illustrate another preferred embodiment, Fig. 4 shows a blending chest 11, at this place, provides the CO through regulating ₂Inlet stream, 7.Fibre stream was passed through measured value, 12, some performance of on-line determination fibre stream before entering blending chest.Simultaneously, measure some on-line performances that export fibre stream through 13.All obtainable on-line determination values are used to calculate nonobservable variable or state, and 14, use real-time model, 15 and observer or best evaluator (estimator), as Kalman filter (Kalman Filter), 16.

The Kalman filtering observer owing to depend on real-time model 15, therefore is commonly called the observer based on model.The observer such as the neutral net of data-driven do not require real-time model, but require a large amount of pending data.The variable that do not observe of the real-time estimation of observer then, and this information delivered to controller, 18, it handles CO ₂Input, the variable that does not observe is changed to desired setting value.

The present embodiment can be used for the Better Wet-end Control and the optimization in paper mill.When can not The real time measure or when obtaining the expectant control variable, it is intended to control in real time wet end.In alleged observational equation, obtainable on-line determination value 12 is relevant with control variables or state with 13.This observational equation is the part of evaluator or observer 6.This observational equation can be supposed some noises of existence in instrument, and the supposition acquisition rate changes.Like this, the used information of evaluator also can be used obtainable information off-line under the long-time interval.

Observational equation is relevant with observer, and state equation too.State equation is the reformulations of model 15, and itself and observational equation all are fit to a kind of algorithm for estimating.State equation also comprises some model uncertainties, the inaccuracy of its corresponding model.Have multiple algorithm for estimating, but one of the most general algorithm is that (Kalman, R.E.) prediction is best evaluator to Kalman filter.The KF evaluator is a kind of Real-Time Evaluation device, and it uses obtainable measured value and obtainable model to estimate nonobservable variable.The KF observer depends on process model 15, so be called the observer based on model.

If can't obtain model, can use observer such as neutral net, but this requires lot of data based on data.Real-time optimization estimated value 14 becomes controller 18 desired process variables measured values, and it is compared estimated value with user-defined setting value, and goes out to control output valve based on these Error Calculation.This control output valve is sent to linear proportioning valve, and it changes CO ₂Interpolation in reactor.

The invention discloses a kind of at employing interpolation CO ₂Wet end processing in control CO ₂The method of adding.With a kind of paper making compositions and CO ₂In conjunction with, generate a kind of CO that is rich in ₂Paper making compositions.Measure or estimate paper making compositions or be rich in CO ₂At least a electrical property of paper making compositions.Control CO then ₂Interpolation speed so that this at least a electrical property remains in the number range of preliminary election.

In a preferred embodiment, from ZP, CD and ion concentration or any performance suitable, select this electrical property with it, and be selected from flow velocity, CD, ZP, ion concentration, Cy, pH value, electrical conductivity and basicity after measured or paper making compositions or be rich in CO ₂Paper making compositions at least a performance and adopt a kind of model that this electrical property is made estimation.Preferred estimation ZP.Thereby, can or be rich in CO from paper making compositions ₂Paper making compositions, and preferably record the performance of this mensuration from paper making compositions.Similarly, can be paper making compositions or be rich in CO ₂Paper making compositions and be preferably and be rich in CO ₂Paper making compositions estimate the electrical property of this estimation.

In a further preferred embodiment, a kind of at employing interpolation CO ₂Wet end processing in control CO ₂The method of adding comprises more accurate control.With paper making compositions and CO ₂In conjunction with, generate a kind of CO that is rich in ₂Paper making compositions.Measure or estimate at least a performance of this paper making compositions and produce the performance data of paper making compositions.The performance data of paper making compositions is offered first line control unit, and it produces the output component of paper making compositions.

CO is rich in mensuration or estimation ₂At least a performance of paper making compositions, produce and be rich in CO ₂The performance data of paper making compositions.And will be rich in CO ₂The performance data of paper making compositions offer feedback controller, it produces the output component of outlet controller.By the output component of analysis access controller and the output component of control of export device feedback controller is compensated.Control CO ₂Inlet stream, make and be rich in CO ₂At least a property retention of paper making compositions in the number range of preliminary election.

This at least a performance of preferred paper making compositions is selected from flow velocity, ZP, CD, ion concentration, Cy, pH value, electrical conductivity and basicity.Be rich in CO ₂This at least a performance of paper making compositions be preferably selected from ZP, CD and ion concentration, and ZP most preferably.Most preferably, this elder generation's line control unit comprises feedforward controller.

In the replacement embodiment of the present embodiment, this elder generation's line control unit comprises feedforward controller, and this feedforward controller uses PREDICTIVE CONTROL or reasoning control.

In another preferred embodiment still, a kind ofly add CO adopting ₂Wet end processing in control CO ₂The method of adding, it comprises slightly different control.With paper making compositions and CO ₂In conjunction with, generate a kind of CO that is rich in ₂Paper making compositions.This paper making compositions of on-line determination is selected from least a performance of flow velocity, ZP, CD, ion concentration, Cy, pH value, electrical conductivity and basicity, produces the performance data of paper making compositions, and sends data to observer.On-line determination is rich in CO ₂At least a performance paper making compositions, that be selected from flow velocity, ZP, CD, ion concentration, Cy, pH value, electrical conductivity and basicity, and produce and to be rich in CO ₂The performance data of paper making compositions.These data are offered observer, and it produces this and is rich in CO ₂The electrical property of at least a estimation of paper making compositions, it is selected from ZP, CD and ion concentration.This observer to controller transmit paper making compositions performance data, be rich in CO ₂The performance data of paper making compositions and the electrical performance data of estimation, and this controller control CO ₂Inlet stream, to keep being rich in CO ₂At least a electrical property of paper making compositions in the number range of preliminary election.

Observer can comprise a kind of model, and the inaccuracy that can bring by analytical model and by the error expected in the assay determination, further improves the electrical performance data of this estimation.The performance data of this paper making compositions, this is rich in CO ₂The performance data of paper making compositions and the electrical performance data of this estimation can incorporate in the software sensors.Can use the electrical performance data evaluation setting value of estimation and carry out real-time closed-loop control.

Note, in all above-mentioned embodiments, by ad-hoc location injection CO in this process ₂, utilizing CO ₂In the wet end processing of chemical property, can control CO in the control wet end ₂Injection so that give liquid or performance that fiber is desired, the two all is present in the paper making compositions.Utilize the performance of liquid or fiber, and the disturbance of process of inhibition can obtain these expected performances.

For example, can directly in fibre stream or fiberless liquid, inject CO at some at least points of wet end ₂, this liquid subsequently can with mixed with fibers.Alternatively, can be with CO ₂Be expelled in the groove.If determined off-line, if then manually control injection is or on-line determination, then injection automatically.

Manually the off-line sample is taken out in control, from these sample determination performances.Then based on some scheme manual adjustments CO ₂These schemes of manually controlling are based on CO ₂The linear or nonlinear recurrence of multivariable between injection, measured value and the desired setting value.

Automatically control can be feedback, feedforward, its combination or first line control unit.FEEDBACK CONTROL is in fibre stream and be rich in CO ₂Stream (pure CO ₂Or and CO ₂The water that mixes) carry out at least once the mensuration of this process after mixing.Feedforward is at injection CO ₂(pure CO ₂Or and CO ₂The water that mixes) carry out at least once the mensuration of this process before, and before process is affected, regulate CO ₂The combination of elder generation's line control unit or above feedforward and feedback, or be such as but not limited to following first line control unit: cascade control, Self Adaptive Control, optimum control, powerful control, nerve network controller, fuzzy control, Model Predictive Control etc.The adjustable joint of elder generation's line control unit CO ₂,, suppress any disturbance simultaneously or the use of chemical addition agent in the wet end minimized to keep desired chemical property in the wet end.

In wet end, pass through CO ₂The chemical property that controller keeps non-exclusively is: ZP (zeta potential), CD (electric charge demand), pH value, ion concentration etc.Can be by CO ₂The disturbance that controller suppresses non-exclusively is: the variation of the variation of waste paper recirculation, inlet electric charge demand, the variation of inlet ZP, pH value, temperature, electrical conductivity etc.CO ₂The performance of employed this technology of controller can be intrinsic, also can be extension.Intrinsic information can be at least but be not limited to following a kind of: pH value, electrical conductivity, CD, ZP, ion concentration etc.Extension information can be at least but be not limited to: flow velocity, volume etc.

The used on-line performance of controller can be that measure or undeterminable.The performance of measuring is at least but is not limited to following a kind of: pH value, electrical conductivity, temperature, flow velocity, CD, ZP etc.Undeterminable or unobservable performance is at least but is not limited to following a kind of: CD, ZP, ion concentration etc.But undeterminable or unobservable performance On-line Estimation, and be used the CO of observer or evaluator ₂Controller utilizes.This observer or evaluator can be predicted undeterminable or unobservable performance based on the algorithm that provides as Kalman filter or neutral net.The procedural knowledge that observer uses comes from basic model, empirical model or inspires model.

What will further note is that in the above-described embodiment, ion concentration can be from [H ⁺], [OH ^-], [Ca+2], [Na+], [HCO ₃-], [CO ₃--] etc.

In addition, in the above-described embodiment, have at least three kinds of technology to be used for controlling automatically: to utilize the performance of paper making compositions, utilization to be rich in CO ₂Paper making compositions in performance and utilize both performance.For first kind of situation, to wish to use feedforward, it can be typical FEEDFORWARD CONTROL or reasoning control or PREDICTIVE CONTROL.Under second kind of situation, preferential is typical FEEDBACK CONTROL, when using Self Adaptive Control, Model Predictive Control, powerful control, optimum control, ANN (Artificial Neural Network) Control, fuzzy control, dynamic matrix control etc.For the third situation, can adopt the combination of two kinds of technology.

Although in aforementioned specification of the present invention, described its some preferred embodiment, and, elaborate many details in order to illustrate.But, it should be apparent to those skilled in the art that the present invention is easy to additional embodiment, and certain details of describing among the application can be done sizable modification with not deviating from basic principle of the present invention.

Claims (12)

1. one kind is being adopted interpolation CO ₂Wet end processing in control CO ₂The method of adding, this method comprises in combination:

With paper making compositions and CO ₂In conjunction with, generate a kind of CO that is rich in ₂Paper making compositions;

At least a performance of this paper making compositions of on-line determination, this at least a performance is selected from flow velocity, ZP, CD, ion concentration, Cy, pH value, electrical conductivity and basicity, produce the performance data of paper making compositions, and the performance data of this paper making compositions is offered observer;

This is rich in CO on-line determination ₂At least a performance of paper making compositions, this at least a performance is selected from flow velocity, ZP, CD, ion concentration, Cy, pH value, electrical conductivity and basicity, produces and is rich in CO ₂The performance data of paper making compositions, and this is rich in CO ₂The performance data of paper making compositions offer this observer;

The performance data that this observer receives this paper making compositions is rich in CO with this ₂The performance data of paper making compositions, and produce this and be rich in CO ₂The electrical property of at least a estimation of paper making compositions, the electrical property of this at least a estimation is selected from ZP, CD and ion concentration, this observer is that this at least a electrical property produces the electrical performance data of estimating, and with the performance data of this paper making compositions, this is rich in CO ₂The performance data of paper making compositions and the electrical performance data of this estimation be transferred to controller, described observer comprises a model; With

This controller control CO ₂Inlet stream, make this be rich in CO ₂At least a electrical property of paper making compositions remain in the number range of preliminary election.

2. the method for claim 1, the performance data of this paper making compositions, this is rich in CO ₂The performance data of paper making compositions and the electrical performance data of this estimation be merged in the software sensors.

3. the method for claim 1, the inaccuracy that this observer brings by analytical model is further improved the electrical performance data of this estimation.

4. the method for claim 1, this observer further improves the electrical performance data of this estimation by the error expected in the assay determination.

5. inaccuracy that the method for claim 1, this observer are brought by analytical model and the electrical performance data of further improving this estimation by the error expected in the assay determination.

6. the method for claim 2, the inaccuracy that this observer brings by analytical model is further improved the electrical performance data of this estimation.

7. the method for claim 2, this observer further improves the electrical performance data of this estimation by the error expected in the assay determination.

8. inaccuracy that the method for claim 2, this observer are brought by analytical model and the electrical performance data of further improving this estimation by the error expected in the assay determination.

9. the method for claim 1 further comprises the electrical performance data of using this estimation, to estimate setting value and to implement real-time closed-loop control.

10. the method for claim 2 further comprises the electrical performance data of using this estimation, to estimate setting value and to implement real-time closed-loop control.

11. the method for claim 8 further comprises the electrical performance data of using this estimation, to estimate setting value and to implement real-time closed-loop control.

12. one kind is being adopted interpolation CO ₂Wet end processing in control CO ₂The method of adding, this method comprises in combination:

With paper making compositions and CO ₂In conjunction with, generate a kind of CO that is rich in ₂Paper making compositions;

At least a performance of this paper making compositions of on-line determination, this at least a performance is selected from flow velocity, ZP, CD, ion concentration, Cy, pH value, electrical conductivity and basicity, produce the performance data of paper making compositions, and the performance data of this paper making compositions is offered observer;

This is rich in CO on-line determination ₂At least a performance of paper making compositions, this at least a performance is selected from flow velocity, ZP, CD, ion concentration, Cy, pH value, electrical conductivity and basicity, produces to be rich in CO ₂The performance data of paper making compositions, and this is rich in CO ₂The performance data of paper making compositions offer this observer;

The performance data that this observer receives this paper making compositions is rich in CO with this ₂The performance data of paper making compositions, and use a model and produce this and be rich in CO ₂The electrical property of at least a estimation of paper making compositions, the electrical property of this at least a estimation is selected from ZP, CD and ion concentration;

This observer improves the estimation to this at least a electrical property by the inaccuracy analyzing this model and bring with by the error expected in the assay determination;

This observer with the performance data of this paper making compositions, this is rich in CO ₂The performance data of paper making compositions and the electrical performance data of this improved estimation flow to controller; With

This controller control CO ₂Inlet stream, make this be rich in CO ₂At least a electrical property of paper making compositions remain in the number range of preliminary election.

Images