US20040134630A1

US20040134630A1 - Method of adjusting pulp washing process and determining efficiency

Info

Publication number: US20040134630A1
Application number: US10/479,951
Authority: US
Inventors: Juri Lahtinen; Mikko Ulvinen
Original assignee: Metso Automation Oy
Current assignee: Valmet Automation Oy
Priority date: 2001-06-21
Filing date: 2002-06-06
Publication date: 2004-07-15
Also published as: FI20011336A0; EP1425466A1; WO2003000984A1; CA2450884A1

Abstract

A method of adjusting a pulp washing process and measuring efficiency. The methods comprise modelling the maximum washing capacity of the washing process, selecting the variables of the washing process to be measured, measuring the variables of the washing process, and adjusting the washing process or determining the efficiency of the washing process by means of the modelling and the measurements.

Description

The invention relates to a method of adjusting a pulp washing process.

The invention further relates to a method of determining the efficiency of a pulp washing process.

In the manufacture of paper pulp, pulp cooking is followed by the step of brown pulp washing. The purpose of the washing is to separate the liquid that comes with the pulp, i.e. so-called base liquor, from the pulp as accurately as possible. The base liquor contains dry solids made up of wood that has dissolved during cooking and cooking chemicals, i.e. so-called waste liquor. During washing, the aim is to separate the base liquor from the pulp flow so that after the washing the brown pulp is as pure as possible for further processing, retaining, however, the base liquor as little diluted as possible. The base liquor separated from the brown pulp is conveyed to be further processed for chemical recovery, wherein the cooking chemicals are recovered to be reused after regeneration and the dissolved wood is burned as a fuel. In the present application, brown pulp refers to unbleached pulp. In comparison with pulp washing, the special characteristics of brown pulp washing include e.g. several factors decreasing the washing potential, such as undecomposable pulp fractions, soap, inorganic material and foam.

The operation of the washers used for washing brown pulp is based either on displacement washing or on dilution-extraction washing. In the displacement washing, cleaner washing water is conveyed through dirty, standard-consistency stock such that the cleaner liquid pushes the dirtier liquid out of the pulp pad. In drum-type washers, the displacement washing is implemented as filter washing wherein filtering is always carried out while subjected to some kind of pressure, and the pressure or suction caused by a fluid column formed during the filtering, or the pressure or suction obtained using a pump, pushes the liquid in the stock through the holes in the drum serving as a filter, which means that the pulp precipitates onto the drum. The brown pulp to be washed in pressure diffusers and atmospheric washing diffusers and the washing liquid are conveyed to opposite ends of a diffuser. The pulp moves between screen rings. The washing liquid is fed into the pulp flow, and the displaced liquid is removed through the screen rings, into a filtrate tank. The washed pulp is removed from the diffuser by means of a specific doctor. During the washing, the entire screen ring system moves cyclically up and down. The screen surface is cleaned by the counter-motion against the direction of travel of the pulp.

In the dilution-extraction washing, stock is diluted by using cleaner washing liquor, whereafter brown pulp is precipitated. This reduces the concentration of the liquor in the stock. Press washers, which are also often provided with a displacement washing step, are based on this operating principle. Various drum-type washing screens and press washers as well as diffusers are known per se to one skilled in the art; therefore, the structure and operating principle thereof will not be discussed in closer detail herein.

A modern brown pulp washing line comprises the above-disclosed washers, wherein a plurality of such washers have been arranged in series. In practice, the washers are coupled in order to enable a perfect counter-current washing to be achieved. In other words, the cleanest washing liquid, which can be warm water or adequately clean, recycled water fraction from a paper mill, usually secondary condensate of an evaporating plant, is conveyed to the last washing step. The filtrate is discharged to become the washing liquid in the previous washing step, and the most concentrated liquor is removed through a digester house to the evaporating plant.

The operation of the washing line devices is usually controlled by a device-specific adjusting system, which is practically always an automatic system. The aim of the adjusting system is to adjust the operation of a washing line and possibly of other washing-dependent functions to be optimal economically. Other washing-dependent functions mainly include a screening department and an evaporating plant. In order to achieve this aim, the washing efficiency is to be determined on the basis of on-line measurements.

Several models exist for describing pulp washing and determining washing efficiency, some of the most important including: washing loss, dilution factor, displacement ratio and Norden efficiency factor.

Washing loss describes the amount of dirt that is transferred together with the pulp to the next washing step. It is presented as a Na ₂SO₄content in post-washing pulp to indicate inorganic washing loss remaining in the stock. A COD (chemical oxygen demand) value, which describes the amount of organic material remaining in the pulp better than the Na₂SO₄content, is also used. The only known possible way to measure washing loss on-line is to measure conductivity. This, however, gives very unreliable results since measuring devices get dirty with time, which causes drifting errors. In practice, washing loss can only be determined by laboratory analyses, i.e. in practice it cannot be applied to optimization or on-line adjustment of washing.

A dilution factor represents the amount of washing liquid used per a certain number of washings. In practice it only describes the efficiency of the evaporating plant, revealing very little about the washing efficiency since the washing potential of the washing liquid varies. The accuracy of the model is thus poor; therefore, it is not sensible to apply the model to the optimization or on-line adjustment of washing.

A displacement ratio describes the amount of dissolved material removed from pulp during washing. However, the displacement ratio cannot be reliably measured on-line; therefore, no reliable results will be obtained from adjustments based thereon.

A Norden efficiency factor indicates the correspondence between the number of perfect dilution-extraction processes and the real process under examination. The Norden efficiency factor can be used for adjusting, but the result is a complex multivariable model which, in order to operate, requires a highly complex adjusting system and detailed information on the washing process. These, however, are not generally available, so extensive and expensive investments in adjustment and measuring devices are required in order to apply the model.

Other methods for adjusting pulp washing are also known. For example, the solution disclosed in U.S. Pat. No. 4,732,651 comprises generating static models by means of certain variables and on-line measuring e.g. the dilution factor. The effect of a change in a variable on the final washing result is predicted on the basis of the models. However, the method does not work if something deviant and unexpected occurs during the washing process. Furthermore, the method only suits a drum washer. Further, in order to operate, the method requires an unexceptionally large number of measuring devices; in practice, no washing line has been implemented with such a number of measuring devices.

The method disclosed in U.S. Pat. No. 5,282,131 aims to avoid the problems caused by long delay times in the adjustment of washing. The publication discloses a method for predicting the amount of dry solids to be conveyed to the evaporating plant. The method can only be applied to drum filter washers. The disclosed method does not work with the existing washers since the technical properties of the equipment restrict the range of the amount of clean washing water to be fed into a washing line to be so narrow that it cannot be adjusted in the extent required by the method. Furthermore, the model does not work if something unexpected occurs during the washing process.

In view of the above, it is problematic to determine the washing efficiency on the basis of on-line measurements since myriad models have to be applied whose field of application has been limited to a specific application and which cannot be successfully used in other parts of the washing line. The directly-measurable variables of the known methods are not very good at describing the efficiency of a washing process, and laboratory analyses must also be used. The known methods do not take into account the role the tanks of a washing line, such as pulp and filtrate tanks, play in a washing event and the washing efficiency, which is a serious drawback as far as the modern washing line thinking is concerned.

An object of the present invention is to provide a novel and improved method of adjusting and measuring a brown pulp washing process.

The method of the invention for adjusting a pulp washing process is characterized by

modelling the maximum washing capacity of the washing process

selecting the variables of the washing process to be measured,

measuring the variables of the washing process, and

adjusting the washing process by means of the modelling and the measurements.

Furthermore, the method of the invention for determining the efficiency of a pulp washing process is characterized by

modelling the maximum washing capacity of the washing process

selecting the variables of the washing process to be measured,

measuring the variables of the washing process, and

determining the efficiency of the washing process by means of the modelling and the measurements.

The idea underlying the invention is that the quality of a washing process is determined by determining its maximum washing capacity. Furthermore, the idea underlying a preferred embodiment of the invention is that the efficiency of a washing process is determined on-line. Further, the idea underlying a second preferred embodiment of the invention is that the maximum washing capacity of a washing process is described as its washing potential in diffusion. Furthermore, the idea underlying a third preferred embodiment of the invention is that the washing potential in diffusion is determined in the following manner:

\begin{matrix} Q_{diff} = k \int \sum_{i = 1}^{n} (x_{i} \nabla c_{i} (t)) e^{- yt} e^{zT (t)} \partial t & (1) \end{matrix}

wherein Q _diff=washing potential in diffusion, k=scaling coefficient, c_i=variables describing concentrations, change in conductivity or in COD or in consistencies, t=time, T=temperature; x, y and z are weighting coefficients and constants. Furthermore, the idea underlying a fourth preferred embodiment of the invention comprises measuring the washing process of an entire washing line.

An advantage of the invention is that it enables the efficiency of substantially all known washers to be monitored, including atmospheric diffusers, pressure diffusers, digester washings, press-type washers, drum washers, wire washers and centrifuge-type washers. A slow diffusion taking place in the pulp tanks of a washing line can also be taken into account. The method of the invention is applicable both to displacement and dilution washings. The implementation of the method in already existing washers and washing lines does not require expensive changes in the measuring equipment thereof, but only the variables that can be measured or determined from the process can be selected to be used. The method can be applied to any commonly used washing process, i.e. a washer or a washing line. A further advantage of the method of the invention is that it comprises no mathematically exact formulas but only some basic rules, according to which e.g. the delay time and concentration differences should be as extensive as possible. Consequently, it is easy to optimize the washing result.

The invention will be described in closer detail in the accompanying examples and drawings, in which [0030]
FIG. 1 is a schematic and highly simplified view showing a washer to which a method of the invention has been applied, and [0031]
FIG. 2 is a schematic and highly simplified view showing an adjustment event of the invention.[0032]

EXAMPLE 1

The operation of a single drum-[0033] type washer 1 of a washing line is optimized. Pulp MF_infrom the previous step is fed into the washer 1, and washing liquid WL_infrom the next step. Correspondingly, pulp MF_outis conveyed from the washer 1 to the next step, and washing liquid WI_outto the previous step. The process taking place in the washer 1 can be measured for the following variables:
1. amount of pulp flow to the washer, [0034]
2. pulp consistency to the washer [0035]
3. amount of pulp flow from the washer [0036]
4. amount of washing liquid to the washer [0037]
5. conductivity of the washing liquid to be conveyed to the washer [0038]
6. conductivity of the washing liquid to be conveyed from the washer [0039]
7. rotation speed of the drum of the washer [0040]
8. temperature of the washing process [0041]
In a schematic and highly simplified manner, the figure shows a situation according to Example 1. The variables of the particular single washer of the washing line can be affected in the following manner: [0042]
1. amount of pulp flow to the washer cannot be affected since this is determined by production, [0043]
2. pulp consistency to the washer: can be affected, [0044]
3. amount of pulp flow from the washer: cannot be affected since this is determined by the pulp flow to be fed, [0045]
4. amount of washing liquid to the washer: cannot be affected since this is determined according to the washing factor of the entire washing line, [0046]
5. conductivity of the washing liquid to be conveyed to the washer: cannot be affected, [0047]
6. conductivity of the washing liquid to be conveyed from the washer: can be affected indirectly through the rotation speed of the drum, [0048]
7. movement speed of the washing event, i.e. presently the rotation speed of the drum: can be affected, [0049]
8. temperature of the washing process: cannot be affected since determined by other process temperatures. [0050]
Selecting the movement speed to be the only variable to be affected, the movement speed being inversely proportional to time: [0051] $\begin{matrix} Q_{diff} = k \int \sum_{i = 1}^{n} (x_{i} \nabla c_{i} (t)) e^{- yt} e^{zT (t)} \partial t & (1) \end{matrix}$
Assume that the concentration difference and the temperature are constants with respect to time, the following form being then derived from Formula 1: [0052] $\begin{matrix} Q_{diff} = \sum_{i = 1}^{n} (x_{i} \nabla c_{i}) \frac{k}{- y} e^{- yt} e^{zT} + c & (2) \end{matrix}$
When t=0, Q[0053] _diff=0, which gives integration constant c the value: $\begin{matrix} 0 = \sum_{i = 1}^{n} (x_{i} \nabla c_{i}) \frac{k}{- y} e^{- y \cdot 0} e^{zT} + c & (3) \end{matrix}$
The final form-obtained is: [0054] $\begin{matrix} Q_{diff} = \sum_{i = 1}^{n} (x_{i} \nabla c_{i}) \frac{k}{y} e^{zT} (1 - e^{- yt}) & (4) \end{matrix}$
The remaining simplification shows that the washing potential in diffusion (Q[0055] _diff) is proportional to the delay time of the pulp on the drum. Now that the movement speed of the washer is known, it can be inferred that the lower the movement speed, the higher the washing potential of the washer. When the movement speed is decreased, a saturation point is eventually reached, after which the washing potential in diffusion no longer substantially increases even if the movement speed were even further decreased.
The washing potential in diffusion describes the washing efficiency under the prevailing conditions and which washing factors affect the washing efficiency. The washing efficiency is described by indicating the maximum possible theoretical washing efficiency that can be obtained. The higher the washing potential in diffusion, the more efficient the washing. The washing potential describes the maximum washing capacity of a washing liquor. It depends e.g. on the electric conductivity of the washing liquor. Diffusion refers to the movement of substances due to the random movement of molecules. The washing potential in diffusion describes the movement of liquid molecules between the pulp to be washed under examination and the washing liquid, wherein the liquid molecules in the pulp fibres are replaced by other liquid molecules. The washing potential in diffusion is based on a completely new approach to a washing process wherein washing is carried out by levelling the concentrations of the components to be removed from the pulp to be washed in the washing medium and in the pulp to be washed. The driving force of the washing is the concentration difference between the components. The basics of the approach can be further led to the second law of thermodynamics. [0056]
The method of the invention for determining and adjusting the efficiency of a washing process differs from the known methods in that the method of the invention allows only directly-measurable variables, such as flows, temperatures, consistencies and filtrate conductivity, to be used. It is also irrelevant which variables are to be or can be used or what the measuring devices of the washing equipment are since the washing potential in diffusion is capable of utilizing all washing variables. The washing potential in diffusion is capable of describing all washing processes in an unbiased manner, which is impossible in the prior art solutions. Norden efficiency factors, for example, do not describe washing in an unbiased manner but, according to these factors, displacement washing is always more efficient than dilution washing, which is not true. [0057]

EXAMPLE 2

An entire brown pulp washing line is adjusted and optimized. As is well known to one skilled in the art, washing is carried out as dictated by other partial processes in a pulp manufacturing process. The arrangement of a brown pulp washing line is determined on the basis of the pulp to be washed and the manner in which it is cooked. Drum washers and press washers can be used for washing brown pulp obtained from batch cooking. The washing line further comprises pulp and filtrate tanks that serve as intermediate storages and buffers between cooking and washing. The volume of the tanks preferably enables substantially continuous washing. [0058]
Typically, at least one diffuser is first applied to continuous cooking and, subsequently, other washer types. [0059]
The measurable and adjustable variables of the exemplary washing line are: [0060]
1. conductivity of the washing liquid to be conveyed to the washing line [0061]
2. conductivity of the washing liquid to be conveyed from the washing line [0062]
3. pulp input consistency to the washing line [0063]
4. pulp output consistency from the washing line [0064]
5. delay time of pulp in the washing process of the entire line [0065]
Assume that the temperature in the washing line is constant, and the average of the concentration differences of the pulp passing-through-time of the entire washing line is used as the concentration difference. The washing potential in diffusion can now be modified in the same manner as has already been shown in connection with Example 1: [0066] $\begin{matrix} Q_{diff} = k \int \sum_{i = 1}^{n} (x_{i} \nabla c_{i} (t)) e^{- yt} e^{zT (t)} \partial t & (1) \end{matrix}$
which is further modified to be of the following form: [0067] $\begin{matrix} Q_{diff} = \sum_{i = 1}^{n} (x_{i} \nabla c_{i}) \frac{k}{y} e^{zT} (1 - e^{- yt}) & (4) \end{matrix}$
wherein k=scaling coefficient, x[0068] _i=weighting coefficient of concentration gradient i, y=inverse of time constant, z=weighting coefficient. It can be inferred that the washing of the washing line becomes more efficient and all possible efficiency of the washing line becomes available by reducing the conductivity of the pulp to be removed as well as by increasing the change in consistency and the delay of pulp on the washing line. A model based on a dilution factor, for example, would have resulted in the conclusion that the more washing liquid used, the better the washing result. This is true to a certain extent but, at the same time, the evaporating costs of the evaporation plant rise exponentially while the washing result practically no longer improves. On the other hand, when appropriately weighted, the washing potential in diffusion also takes the costs incurred by the process of evaporating liquid into account. The weighting can be carried out by binding the weighting coefficients and/or constants together by a formula.
The washing potential in diffusion can be utilized for optimizing the washing process e.g. by calculating an estimate of the washing potential in diffusion of the entire washing process after half the washing time t has elapsed. This enables adjustment procedures to be carried out in order to optimize the process. [0069]
The efficiency of the washing of a washing line can also be determined by calculating washer- or washing-step-specific washing potential in diffusion separately for each washer or washing step in the line. In other words, the washing line is divided into successive steps. Integration of a step is stopped at a point in time when the particular step ends, and the calculation is continued by integrating the next step. Finally, all steps are summed up, which gives the washing potential in diffusion of the entire washing line. Naturally, the calculation has to take into account the pulp tanks on the line as well as other process steps that usually even out concentration differences. [0070]
The washing taking place in a single washer or washing step can also be divided into two or more steps. The washing potential in diffusion of each step is determined separately, and the washing potential in diffusion of the entire washer or washing step is obtained by summing up the washing potentials in diffusion of the steps. The washing taking place in a TwinRoll press washer, for example, can be divided into three parts: the first step before the nip of a press, the second step the nip, and the third step the pulp conveyor screw after the nip. [0071]
The washing potential in diffusion is an exponential model; therefore, it takes into account the dynamics of a process. It also takes into account the way in which a change in different variables affects the washing efficiency. [0072]
In addition to so-called fast diffusion taking place in the actual washers, the washing potential in diffusion also takes into account so-called slow diffusion, i.e. the washing efficiency of the various pulp tanks in a washing line, which is a most essential feature as far as the modern washing line thinking is concerned. In a filtrate tank, for example, particles bind efficiently to black liquor whereas in a pulp tank, pulp has a very long delay time. These factors cannot be described by the prior art models. For example, the washing efficiency of a filtrate tank can be determined by measuring the consistency of the pulp conveyed to the tank and the pulp conveyed therefrom. This gives a concentration change. When also time is taken into account, the washing potential in diffusion can be calculated for the particular tank. [0073]
The washing potential in diffusion can be applied for describing the efficiency of different washers and comparing changes in efficiency with each other. It can be applied as a so-called Soft Sensor for describing the operation of washers. A Soft Sensor is a model of a non-measurable variable, which is based on other, measurable variables. It can also consist of modification of measured information into a more usable form. Since there are several characteristics of brown pulp that cannot be measured directly, Soft Sensors are very useful in this connection. [0074]
FIG. 2 is a schematic and highly simplified view showing an adjustment event of the invention. A [0075] washing line 2 to be adjusted comprises one or more washers. The washers and/or washing liquid of the washing line 2 is/are measured for the above-mentioned variables, for example. Measurement results 3 are fed into an adjusting device 4 adjusting the operation of the washing line. The adjusting device 4 comprises a model describing the maximum washing capacity of the washing line 2. Employing the model, the adjusting device 4 processes the measurement results 3 and determines an adjustment message 5 on the basis of these results. The adjustment message 5 is transmitted to the washing line wherein actuators carry out the procedures according to the adjustment message 5.
The drawings and the related description are only intended to illustrate the idea of the invention. In its details, the invention may vary within the scope of the claims. The method is thus also applicable to bleaching. The ion quantities of pulp are then measured prior to and after a washer, and the material balance of the process is calculated. [0076]

Claims

1. A method of adjusting a pulp washing process, characterized by

modelling the maximum washing capacity of the washing process by a dynamic model,

selecting the variables of the washing process to be measured,

measuring the variables of the washing process, and

adjusting the washing process by means of the modelling and the measurements, wherein the maximum washing capacity of the washing process is described by its washing potential in diffusion and wherein the washing potential in diffusion is determined in the following manner:

\begin{matrix} Q_{diff} = k \int \sum_{i = 1}^{n} (x_{i} \nabla c_{i} (t)) e^{- yt} e^{zT (t)} \partial t & (1) \end{matrix}

wherein Q_diff=washing potential in diffusion, k=scaling coefficient, c_i=variables describing concentrations, change in conductivity or in cod or in consistencies, t=time, t=temperature while x, y and z are weighting coefficients and constants:

2. A method as claimed in claim 1, characterized by optimizing the washing process on the basis of the modelling and the measurements.

3. A method as claimed in claim 1 or 2, characterized by adjusting the operation of the washing process on-line.

4. A method as claimed in any one of the preceding claims, characterized in that the pulp to be washed is brown pulp.

5. A method as claimed in any one of claims 1 to 3, characterized in that the washing process takes place in a pulp bleaching process.

6. A method as claimed in any one of the preceding claims, characterized in that the washing process to be adjusted comprises an entire washing line.

7. A method as claimed in any one of claims 1 to 5, characterized in that the washing process to be adjusted takes place in a single washer.

8. A method as claimed in any one of the preceding claims, characterized in that the modelling of the maximum washing capacity of the washing process is applied as a Soft Sensor.

9. A method of determining the efficiency of a pulp washing process, characterized by

modelling the maximum washing capacity of the washing process,

selecting the variables of the washing process to be measured,

measuring the variables of the washing process, and

determining the efficiency of the washing process by means of the modelling and the measurements, wherein the efficiency of the washing process is described by a dynamic model and by its washing potential in diffusion and wherein the washing potential in diffusion is determined in the following manner:

\begin{matrix} Q_{diff} = k \int \sum_{i = 1}^{n} (x_{i} \nabla c_{i} (t)) e^{- yt} e^{zT (t)} \partial t & (1) \end{matrix}

wherein Q_diff=washing potential in diffusion, k=scaling coefficient, c_i=variables describing concentrations, change in conductivity or in COD or in consistencies, t=time, T=temperature while x, y and z are weighting coefficients and constants.

10. A method as claimed in claim 9, characterized by optimizing the washing process on the basis of the determination of the efficiency of the washing result.

11. A method as claimed in claim 9 or 10, characterized by determining the efficiency of the washing process on-line.

12. A method as claimed in any one of claims 9 to 11, characterized in that the pulp to be washed is brown pulp.

13. A method as claimed in any one of claims 9 to 11, characterized in that the washing process takes place in a pulp bleaching process.

14. A method as claimed in any one of claims 9 to 13, characterized by measuring the efficiency of an entire washing line.

15. A method as claimed in any one of claims 9 to 13, characterized by measuring the efficiency of a single washer.

16. A method as claimed in any one of claims 9 to 15, characterized by applying the modelling of the maximum washing capacity of the washing process as a Soft Sensor.