FI92225C - Procedure for using a support model - Google Patents

Description

92225

Method for using a cooking template

The invention relates to a method for using a cooking model in connection with a computer-aided cooking process, such as sulphate, sulphite, batch cooking, cooking or the like, wherein the cooking process is controlled, at least initially, by a calculated cooking model using a predetermined Kpa , the relationship between the parameter describing the alkali concentration and the H-factor H, which cooking model is based on the so-called To Hatton's Kappa model, which binds the Kappa number K, the effective alkali EA and the H-factor H, 15 having the form: K = A - B x [log (H) x (EA) n], where the first factor A, the second factor B and the third factor 20 n are factors depending on wood species, wood quality, mass and cooking kinetics, etc.

The Kappa model was developed by J.V. Hatton (Department of 25 the Environment, Canadian Forestry Service, Western

Forest Laboratory, Vancouver) -70s. In this case, on the basis of the experimentally found relationship between the Kappa number K, the effective alkali EA and the H-factor (Vroom cooking factor), the above equation has been constructed.

30 The H-factor in the formula describes the progression of the cooking reactions throughout the cooking using the cooking time and the cooking temperature, in which case it can be expressed in the form H = exp (43.2 - 161l3 / T) dt, where T = cooking temperature 35 in Kelvins.

Hatton's Kappa model is useful for cooking control because it is 2 92225 - experimentally tested, - easy to apply in an automation system,, - binds important cooking factors together, 5 - allows each factor to be calculated separately when other factors are known, - gives a sufficiently accurate result when tuned to factory conditions , - modified, also suitable for calculating the cooking yield 10, and - is applicable to both batch and batch cooking.

However, the introduction of soups 11 in has been limited by the fact that the factors A, B and n 15 shown in the cooking model are constants that strongly depend on wood species, wood quality, mass and cooking kinetics and the like. The values of these constants have been reported in the literature, e.g., for some North American tree species for certain seasons, 20 but the values do not correspond to tree species or climatic conditions elsewhere in the world. Likewise, local factory conditions have a significant effect on the values of the coefficients. In this case, the practical problem has been the difficult tuning of the soup model for the wood species to be treated 25 in the factory conditions and process in question. Disadvantages of tuning the Hatton model have been reported by e.g. Esa Jutila: A survey of kraft cooking control models, IFAC, 1980). Thus, despite its advantages, the Hatton Kappa model has not previously been more widely used in process control 30 due to its tuning problems. With regard to the prior art, reference is further made to US-5032977, which discloses the use of the relationship between the alkali parameter of the body and the H-factor in the cooking model to achieve the desired body number.

The object of the method according to the present invention is to provide a decisive improvement over the above-mentioned drawbacks and thus to increase the state of the art 3,92225. To achieve this purpose, the method according to the invention is mainly characterized in that the cooking pattern is used in the form: 5 K - AI + B1 X [log (H) x (AL) n], where the oset parameter AI, the steepness parameter B1 and the excitation parameter n are of wood species , wood quality, mass and cooking kinetics, etc. 10, and where the time parameter AL describes the effective alkali, the cooking model of which at least solves the functional Kappa number The species-specific Kappa number K, H factor H and alkali parameter AL using target values.

15

The most important advantages of the method according to the invention are its simplicity and accuracy, whereby especially in the commissioning phases it is easy to tune the cooking control for each pulp-20 and wood species, whereby it is also possible to maintain the cooking control by operating personnel. In this case, maintenance personnel or the like are no longer required to tune the Kappa control of the cooking control. The cooking model can also be used well in exceptional situations, such as changes in production, changes in production speed, etc., where cooking control is further emphasized. As driving conditions change, the benefits of the cooking model are greatest in the pursuit of better quality control. Thus, the cooking model according to the invention enables the use of the Kappa model in a more versatile and efficient way than the systems currently in use.

Other dependent claims present preferred solutions for applying the method according to the invention.

4,92225

In the following description, the invention is illustrated in detail.

f

The method according to the invention is intended to be used as soups for 1-5 liters in connection with a computer-aided system, such as most commonly a digital automation system, in particular a pulp cooking process such as sulphate, sulphite, batch soup, cooking or the like, at least initially. Hatton

Using the Kappa model of the form: K = A - B x [log (H) x (EA) n], which is used in the method according to the invention in the form: K = AI + B1 x [log (H) x (AL) n], 20 where of fset parameter A1, steepness parameter B1 and excitation parameter n are factors depending on wood species, wood quality, mass and cooking kinetics, etc., and where the alkali parameter AL describes effective alkali. The cooking model 25 shown solves the functional Kappa number Kt by means of known species-specific target values of the kappa number K, H factor and the alkali parameter AL, using a predetermined, at least Kappa number K, alkali parameter AL and H factor influence ratio.

To apply the method, it is preferred to use the alkali parameter in the form of AL, where it describes the effective alkali dose as a percentage of the absolutely dry wood (equivalent to Na 2 <3 or NaOH, Na 2 O = 0.775 x NaOH).

5,92225

To solve the cooking model, it is advantageous to determine the relationship between the Kappa number K, the alkali parameter AL and the H-factor using two sensitivity parameters, where the first sensitivity parameter dH / dK 5 describes the rate of change of the H-factor with respect to the Kappa number K and the second with respect to Chapter K. The first sensitivity parameter dH / dK indicates how many H-factor units are needed to change the Kappa number by one unit when at a given operating point; eg for Nordic softwood at a hardness level of Kappa 36 ± 4, the value is about 67 H-factor units, source:

Manufacture of wood pulp Part II 1. The second sensitivity parameter dAL / dK indicates how many percentage points of the alkali dose are needed to change the Kappa number by one unit at a given operating point; for example, for Nordic softwood at a hardness level of 36 ± 4, the value is about 0.25% alkali, source: Manufacture of wood pulp Part II 1.

20 With the help of the defined sensitivity parameters dH / dK, dAL / dK it is possible to calculate the excitation parameter n from the first equation I: n = [log (e) x AL x dH / dK] / [log (H) x H x dAL / dK], 25 where e = 2,17183, after which it is still possible to calculate the steepness parameter B1 from the second equation II using the calculated excitation parameter n and the determined first sensitivity parameter dH / dK B1 = - [ln (10) x H] / [(AL) nx dH / dK], after which it is still possible to calculate the offset parameter AI from the third equation III on the basis of the calculated steepness parameter B1 and the excitation parameter n 35 and the target value K of the Kappa number: AI = K - B1 x [log (H) x (AL) n] .

6 92225 In this case, the calculated offset parameter AI, the steepness parameter B1 and the excitation parameter n as well as the known i-specific target values of the H-factor and the alkali parameter AL 5 make it possible to solve the functional Kappa number K ^ by placing them in the cooking model:

Kt = Al + B1 x [log (H) x (AL) n].

10 The above equations I, II, III are derived from the cooking model: K = AI + B1 X [log (H) x (AL) n], (1) 15 from which the following equations are obtained at a given operating point: K - dK = AI + B1 [log (H + dH) x (AL) n] (2) K - dK = AI + B1 [log (H) x (AL + dAL) n] (3) 20 by marking the equations derived from the cooking model equal to the alkali dose AL exponent n solved: 25 n = {log [log (H + dH) / log (H)]} / (log [(AL + dAL) / AL]> (4)

When estimating the expression log (H + dH) with the expression: log (H) + (kx dH) / H, where k = log (e), e = 2.7183, 30 equation (4) is simplified to equation I: n = [ log (e) x AL x dH / dK] / [log (H) x H x dAL / dK] (5) (pine mass Kappa level 19, H = 1100 ... 1400, goodness of estimate 35:

Err ^ = 1.91 X 10 “® Sum of squared residuals = 10.0 x 10“ 1 Square of multiple correlation coefficient 7 92225

Sig = 0.0 x 10 "! Significance of fit

Max Err = 1.6 x 10 "* Maximum error)

Equation 5 defined by the steepness parameter B1 of the cooking model can be derived by subtracting from Equation (1) Equation (2) to give B1 = dK / {(AL) nx log [H / (H + dH)]>, (6) 10 which is further simplified by equation II, i.e. B1 = - [ln (10) x H] / [(AL) nx dH / dK] (7) 15 To solve the offset parameter AI, solve Equation 1 with respect to: AI = K - B1 x [log (H) x (AL) n], (8) 20 where the known target value of the Kappa number is used as the Kappa number.

The Kappa model is used to control cooking by species. Sensitivity parameters dAL / dK are simply given for each pulp / wood type 25 to be driven.

and dH / dK as well as the target values K, H, AL, which species-specific values are stored in the recipes of the automation system.

30 The tuning of the model is calculated from Equations I, II and III

automatically either at regular intervals or in the event of a change.

Once the model is in tune, the output values can be used to determine any of the cooking factors when other factors are known.

35

By means of the method according to the invention, it is possible to simulate driving situations by means of an automation system by giving different values of the output variables and 8 92225

by reading the predicted variable values from the system monitors. In this way, for example, in a species change situation, the effects of the new parameter values can be studied before changing the actual process quantities. The 5 quantities to be simulated are typically the required alkali dose AL

* or H-factor. Simulations can be used to assess, for example, in situations of change and limitation, whether the desired quantity and quality of production can be achieved by saving chemicals.

10

A Finnish sulphate batch digester used the method to determine the potential of cooking control to save chemical consumption. By increasing the H-factor, the aim was to increase the cooking time, i.e. to save steam energy. Senna wanted to keep production volume and pulp quality (Kappa level and dispersion) constant. The model was used to calculate that such a measure is possible.

It also turned out that this extension of the cooking time and the corresponding steam savings can also be used as 20 chemical savings. This made it possible to reduce the alkali dose by 7%. After the test run, it was found that the alkali dose in question was able to drive the desired production rate and quality, so that after the test run the new setpoints were introduced continuously.

25

It is clear that the invention is not limited to the application presented above, but can be considerably modified within the framework of the basic idea. Of course, by selecting one or more parameters other than those set forth above, the applicable formulas may differ substantially in appearance from those set forth. Of course, the formulas used in this presentation can also be rearranged in very different forms. For example, on the basis of the known species-specific 35 functional Kappa numbers and alkali parameters, it is possible to solve the operating point H-factor in the form of a cooking model: 92225 9 H = exp [(Kt - AI) / (B1 x (AL) n)], or the functional Kappa number and Based on the H-factor, the alkali parameter of the operating point from the cooked form is: AL = exp {1 / n [log ((Kt - AI) / (B1 x log (H)))]> 4 i

Claims (6)

A method of using a digest model in connection with a digest process, which is controlled by a computer-based system, such as sulphate, suitite, batch, flow depletion or the like, wherein the digest process is initially controlled at least with the aid of a determinable effect relationship between at least one Kappa number (K), an alkali concentration descriptive parameter and an H-factor 10 (H), which starvation model is based on Hatton's Kappa model, using the starvation model in the form: K = Al + 51 * [log (//) * (/ 4L) "], where the offset parameter (A1), the gradient parameter (B1) and the tuning parameter (n) are of the type of wood, wood quality, pulp and digestion kinetics, etc. similarly dependent factors and where the alkali parameter (AL) describes the effective alkali, by means of which the dissolution model dissolves at least the functional Kappa number (Kt) by means of the known woody mean values of the Kappa number (K), the H factor (H) and the alkali parameter (AL), characterized in that the effect relationship between the Kappa number (K), the alkali concentration descriptive alkali parameter (AL) and the H-factor (H) ) is determined by at least one, preferably two, sensitivity parameter, wherein - the first sensitivity parameter (dH / dK) describes the rate of change of the H-factor (H) relative to the Kappa number (K), and 30. the second sensitivity parameter (dAL) / dK) describe is the rate of change of the alkali parameter (AL) relative to the Kappa number (K), * - at least one sensitivity parameter is conducted by the digest equation K = A \ + B \ * ^ and (H) * {AL) n] the terms of the offset parameter (A1), the gradient parameter (B1) and the tuning parameter (n), and 5. the expressions are resolved by the tree-like mean values of the Kappa number (K), the H factor (H) and the alkali parameter (AL). Method according to claim 1, characterized in that, by means of the determined sensitivity parameters (dH / dK, dAL / dK), the tuning parameter (n) is calculated from a first equation (I): n = [log (e) * AL * dH / dK] / [\ og (H) * H * dAL / dK], where e = 2.7183. Method according to claim 1 or 2, characterized in that a gradient parameter (B1) is calculated from a second equation (II) by means of the calculated tuning parameter (n) and the determined first sensitivity parameter (dH / dK). = - [ln (l 0) * //] / [(AL) H * dH / dK \ Method according to any of claims 1-3, characterized in that on the basis of the calculated gradient parameter (B1) and the tuning parameter (n) and the measured value of the Kappa number (K), an offset parameter (A1) is calculated from a third equation ( III): A1 = K-51 * [log (//) * (AL) "]. 30 Method according to any of claims 1-4, characterized in that by means of the calculated offset parameter (A1), the gradation parameter (B1) and the tuning parameter (n) as well as the known tree species measured values of the H factor (H) and the alkali parameter (AL), the functional Kappa number (Kt) is calculated by placing them in the digest model. I: *, l Kt = A \ + B \ * \ log {H) * (AL) * \ 15 92225 4 6. A process according to claim 1, characterized in that the known mean value of the woody alkali parameter (AL) is determined to * describe the effective alkali portion as a percentage calculated from the absolute dry tree (calculated as equivalent Na 2 O or NaOH, Na 2 O = 0.775 x NaOH ). m)