JP2000094497A - Film thickness control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a film thickness control device for converging the thickness profile into the stabilized stationary control state efficiently in a short time. SOLUTION: A film thickness control device is provided with a plurality of operation ends for adjusting the thickness of a die in each of respective given widths provided all over the whole width of the die for molding a film from molten resin, a thickness indicator for measuring the film thickness on respective measuring ends corresponding at least to respective operation ends in the film width direction and a control means for controlling respective operation ends based on the respective thickness measured values measured by the given periods and respective target values set preliminarily. The thickness of the films to be manufactured is controlled to the target values, and the control means is provided with a control simulation means for simulating the control result of the given control operation set based on a process module formed of a numerical formula formed by utilizing the thickness measured values and target values so that the operation ends are controlled by the control output provided by the control simulation means.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for controlling the thickness of a film (also referred to as a sheet).
The die thickness adjusting means for molding a film made of a molten resin comprises a plurality of operation ends for adjusting the thickness by operating the discharge amount for each predetermined width of the die disposed over the entire width of the die,
A film thickness control for controlling a film thickness profile by a control means including a plurality of control loops for controlling the thickness by operating the operation end based on the film thickness detected at least at each measurement point corresponding to the operation end. The present invention relates to a film thickness control device capable of effectively converging a thickness profile to a stable steady state in a short time.

[0002]

2. Description of the Related Art To control the thickness of a sheet-like material, for example, a plastic film in a width direction to a predetermined profile, for example, to uniformly control the film thickness, Japanese Patent Publication No. 6-7590.
No. 6, JP-B-6-75907, JP-B-6-7
As described in Japanese Patent Application Laid-Open No. 5908 and the like, a plurality of operation terminals for controlling the thickness by controlling the discharge amount of a predetermined width disposed over the entire width of a wide die of the extrusion molding device and the casting die for forming the same. More specifically, a multi-unit control loop for controlling each of the thickness adjusting means including operating end units such as a heater and a gap adjusting tool based on the film thickness measured at each corresponding downstream measuring point. Generally, it is based on point control means.

As the multi-point control means, each control loop is independent and outputs a control output obtained by performing P, PI or PID calculation of a well-known control operation on the deviation between the detected thickness and the target value. PID output to thickness adjustment means as operation amount
Control is widely used because of its stable configuration, simple configuration, and easy tuning.

[0004]

The above-described control method using the multi-point control means provides profile control that has practically no problem in a normal operation state in the case of a normal single layer. However, there is a problem that it takes a lot of wasted time to converge to the thickness unevenness of the target quality at the time of starting up or when a large condition change is performed, which takes an enormous amount of time.

Further, when adjusting the profile of each layer of the multilayer film, there is a problem that it takes much time to cause interference and converge between the layers as compared with the case of a single layer.

An object of the present invention is to solve such a problem, and an object of the present invention is to provide a film thickness control device capable of effectively converging a thickness profile to a stable control state in a short time.

[0007]

According to the present invention, as a result of intensive research to achieve the above object, a measured thickness value is input to a process model consisting of a mathematical model obtained through experiments and the like, and a control simulation is performed. It has been found that the thickness profile can be adjusted in a short time by outputting the control output of the simulation result when the target thickness profile is converged to the operation end of the actual process as an operation amount.

That is, the present invention provides a plurality of operation ends for adjusting the thickness of a die, which is provided over the entire width of a die for molding a film from a molten resin, at a predetermined width, and at least the operation ends in the film width direction. A thickness gauge that measures the thickness of the film at each measurement point corresponding to each of, and a control unit that controls each operation terminal from each thickness measurement value measured at a predetermined cycle and each preset target value. A film thickness control device for controlling the thickness of a film to be manufactured to the target value, wherein the control means sets a predetermined value set based on a process model consisting of a mathematical model from the thickness measurement value and the target value. Control simulation means for simulating a control result of the control operation of the control operation means, wherein the operation end is controlled by a control output obtained by the control simulation means. That has been made is the thickness control device for film characterized.

In the present invention, the control simulation means includes a convergence determination means for determining whether or not the calculated control result has converged, and the first simulation value obtained based on the input thickness measurement value and target value. If the control result is not determined to be convergence by the convergence determining means, a procedure for obtaining the second control result using the first control result as the next thickness measurement value and determining the second control result by the convergence determining means is performed by the convergence determining means. It is preferable from the standpoint of data processing, especially online processing, that the operation is repeatedly performed until the convergence is determined, and the control output when the convergence is determined is output to the operation terminal.

In view of this on-line processing, the process model is a static first-order linear model. In particular, a linear interference model in which mutual interference between a plurality of neighboring operation terminals is approximated by a first-order equation using a coefficient of interference as a coefficient. However, it is preferable in that high-speed processing can be performed and the control result is good.

In the case where the film is a multi-layer film having two or more layers, the process model is expanded, and the influence of the operation end of the other layer on the layer is approximated by a linear expression using a process gain as a coefficient. Multi-layer models are preferred for similar reasons. Hereinafter, details of the present invention will be described.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the drawings based on an embodiment in which the present invention is applied to a manufacturing process using a three-layered multilayer film die.

FIG. 1 is a schematic illustration of the basic configuration of an embodiment of the present invention. In FIG. 1, a die 1 is disclosed in
No. 2, Japanese Utility Model Application Laid-Open No. 7-15321, etc., is provided with a plurality of flow paths for laminating a molten resin, a multilayer extrusion die for continuously extruding a plurality of films into a multi-layer film shape. First surface layer / adjacent layer / second layer provided with thickness adjusting means 2 and 3 for each layer
Are formed to form a multilayer film of three layers of the surface layer, and the raw material resin of each layer is melted and extruded from a surface resin extruder and an adjacent layer resin extruder (not shown). It is supplied to a manifold (not shown).
The thickness adjusting means 2 and 3 of each layer have a well-known configuration described later, and have a configuration in which adjustment units for adjusting a predetermined width are arranged side by side over the entire width of the film.

Although the film has a three-layer structure in this embodiment, it may be a single layer or a multilayer film having two or more layers. The structure of the film 5 is 2
In the case of a layer, the resin of each layer may be merged upstream of the die and extruded from the die slit, or may be merged in the die. If the number of layers is three or more, it is better to control the thicknesses of the respective layers so that they are merged in the die 1 for better controllability.

If the multilayer film has three or more layers, the resin for the surface layer is extruded from the slit of the die 1 so as to sandwich the adjacent layer in the die as in this example. The extruded multilayer film 5 is cooled and solidified by the cooling drum 4, stretched by the stretching device 6, and wound up by the winding device 9.
Winded up by.

The total thickness of the stretched film is measured in the width direction of the film by measuring while scanning in the width direction of the film using a radiation transmission type thickness meter 7 that measures the thickness using the transmission of radiation. be able to. The surface layer thickness of the stretched film is measured by a surface layer thickness meter 8 that measures only the thickness of the surface layer while scanning in the width direction of the film as in the radiation transmission type thickness meter 7. At this time, it is preferable that the positions of the radiation transmission type thickness meter 7 and the surface layer thickness meter 8 be scanned synchronously in order to obtain the thickness of the adjacent layer.

The measured thickness data of each layer together with the measured position data in the width direction of the film are input to a controller 10 comprising a computer and fed back. The controller 10 includes an input device for inputting thickness measurement values of the surface layer and all layers of the multilayer film 5 and measurement position data thereof,
An arithmetic unit that calculates each control output such that the difference between the thickness measurement value at each measurement position corresponding to each operation end for each layer and each preset target thickness is zero, and a processing procedure, A memory for storing necessary data such as calculation results, thickness adjusting means 2 and 3 each comprising an operation terminal for adjusting the thickness in accordance with each control output from the arithmetic unit, a monitor, a keyboard,
It consists of an input / output device such as a mouse.

The measured thickness of the adjacent layer is obtained by subtracting the measured value of the thickness of the surface layer from the measured value of the thickness of all the layers at each measurement position by the controller 10.

The operating ends of the thickness adjusting means 2 and 3 may have any structure as long as the thickness of each layer having a predetermined width can be controlled along the die width direction. Can be applied. . However, from the viewpoints of operability, maintainability, simplification of the die structure, and the like, the surface layer thickness adjusting means 2 is preferably a temperature adjusting method of changing the resin flow rate by changing the die lip temperature using a heater. On the other hand, the thickness adjusting means 3 of the adjacent layer is used to change the resin flow rate by changing the opening degree of the flow path in the die by using a heat bolt in terms of operability, maintainability, and simplification of the structure of the die. The bolt system is preferred.

If the multilayer film has a different refractive index between the surface layer and the adjacent layer, reflection occurs at the interface between the two layers. Therefore, the thickness of the surface layer depends on the reflected light at this interface and the reflection on the surface of the surface layer. The thickness can be stably measured even when the thickness is thin by an optical interference type thickness meter that measures the layer thickness using light interference of light, and this film thickness meter was applied to the surface layer thickness meter of this example.

The material of the surface layer of the multilayer film to which the present invention can be applied includes all resins which transmit light and can be extruded with a die. The material of the adjacent layer includes all resins that can be extruded with a die, but when controlled independently of the surface layer, a material having a different refractive index from the surface layer is preferable from the viewpoint of measuring the surface layer thickness. Representative examples of these resins include polyesters such as polyethylene terephthalate and polyethylene naphthalate, and polyethylene. Further, these copolymers and mixtures may be used, or may further contain other additives and the like.

By the way, the arithmetic unit of this embodiment is constituted by the following control means. FIG. 2 is a block diagram of the basic configuration of the thickness control, and FIG. 3 is a flowchart of the configuration of the control means constituting the controller of the basic configuration of FIG.

As shown in FIG. 3, the control means determines whether or not the thickness deviation of each operating end of each layer is within a threshold value determined for each layer from a management value such as thickness unevenness which is a preset product quality target. A convergence determining means for determining whether or not the thickness control has converged to the target thickness. First, at a fixed cycle, a thickness deviation input between the measured thickness value and the target thickness is obtained, and the convergence determination is performed. If the thickness deviation converges within the threshold value, the current control output is maintained, and the same operation amount is continuously output to the operation terminal.

If the convergence determining means determines that the thickness deviation does not converge beyond the threshold value, control is performed by the following control simulation means. That is, as shown in the figure, the control simulation means first uses a control algorithm of a control operation preset in each layer thickness control calculation unit based on the thickness deviation input (in this example, a well-known PID control is used). A control operation is performed to obtain a first control output for each operation end of each layer. The control result is calculated based on a process model in which the actual process of controlling the thickness of the film in FIG. 2 obtained from the first control output by experiment or theory or by using both of them is approximated by a mathematical model. The first control result by simulation, that is, the first simulation result of each layer thickness profile generated by the first control output is obtained. Next, the result of the control simulation means is judged by the convergence judging means.
The second control result is obtained by performing the same calculation as the first control using the second control result as the second thickness measurement, and the second control result is determined by the convergence determining means. When it is determined that the control has not converged, the second control result is used as the third thickness measurement, and this procedure is repeated until the control result of the simulation converges. Then, the control output of the simulation means when the convergence determination means determines that the control result has converged is output to each operation terminal as an operation amount.

As described above, by obtaining the control output by the control simulation means using the process model, the control output that will converge the thickness can be output. You can adjust your profile. Note that in this example, P
Although the control using the ID control has been described, various control operations such as P and PI control and shortest time control can be applied. In addition, the control result of the thickness control of the actual process and the control result of the control simulation unit are determined by the same convergence determination unit. May be used as the state determination means for determining the condition. This method has an advantage that it can support a wider range of processes.

In the present embodiment, the process model used in the control simulation means described above is the one shown in the flowchart of FIG. This process model is an approximation of the multilayer film production process using a simple mathematical model.
It has the following configuration.

First, when the control output for each operating end of each layer obtained by the control operation of the control operation described above is input, the change is obtained and the change is multiplied by the process gain to obtain the control output. The change is converted into a thickness change. In this conversion, the thickness changes △ H1 (n) and △ H2 (n) of the first and second surface layers corresponding to each operation end n and the thickness change △ R (n) of the adjacent layer are caused by a process gain of the interlayer interference. Aij
Was evaluated using the following linear linear multilayer model. Here, △ OH1 (n), △ OH2 (n), and OROR (n) are the operating terminals n of each layer.
Is the change in the control output.

[0028]

数 H1 (n) = A11 · △ OH1 (n) + A1r · △ OR (n) +
A12 △ OH2 (n)) R (n) = Ar1 △ OH1 (n) + Arr △ OR (n) + Ar2 △ OH
2 (n) △ H2 (n) = A21 · △ OH1 (n) + A2r · △ OR (n) + A22 · △ O
H2 (n)

Then, thickness changes H1 (n) and H2 as thickness measurement values corresponding to each operation end n with respect to the film of each layer.
(n), R (n) is a linear interference model by a linear equation of the following equation using the interference rate Bk, n as a coefficient in consideration of the interference between the operation ends in the width direction of the film in each layer and the draw ratio. I asked.

[0030]

## EQU2 ## H1 (n) = B1, n-2 △ H1 (n-2) + B1, n-1 △ H
1 (n-1) + B1, n △ H1 (n) + B1, n + 1 △ H1 (n + 1) + B1,
n + 2 · △ H1 (n + 2) R (n) = Br, n−2 · △ R (n−2) + Br, n−1 · △ R (n−1) + B
r, n · △ R (n) + Br, n + 1 · △ R (n + 1) + Br, n + 2 · △ R
(n + 2) H2 (n) = B2, n-2 △ H2 (n-2) + B2, n-1 △ H2 (n-1) +
B2, n △ H2 (n) + B2, n + 1 △ H2 (n + 1) + B2, n + 2 △
H2 (n + 2)

In the above equation, two operating ends n-2, n-1 and n which are adjacent to the operating end n on the left and right sides, respectively.
It is assumed that there is interference up to +1 and n + 2, and that there is no influence of the operating end farther than that. Note that this number is preferably obtained by an experiment or the like depending on the process.

In an actual process, a time constant and a dead time are required from the time when an output is normally output to the operation terminal to the time when the change is measured. Therefore, as a process model, this dynamic process is also approximated in [] in FIG. Although the mathematical model including the calculation of the time constant and the calculation of the dead time is accurate, but the purpose is to obtain the thickness profile, a static model ignoring this dynamic process was used. As a result, the calculation process was greatly simplified, and a model suitable for online processing was obtained.

The control simulation means calculates the thickness change amount obtained by the above process model from the amount of resin extruded from the die in the actual process becomes substantially constant per time. The uniformization is corrected so that the sum of the thickness change amounts becomes constant. By adding the corrected thickness change to the previous control result, a new simulation result, that is, a thickness output of each layer is obtained. Hereinafter, an example of film formation according to the above-described embodiment will be described.

[0034]

[Example of film formation] A multilayer film having a three-layer structure of first surface layer / adjacent layer (core layer) / second surface layer was formed. The same resin was used for both surface layers. As a polyester material for the surface layer,
12 mmol of potassium acetate based on dicarboxylic acid component
%, A pellet of polyethylene terephthalate having an intrinsic viscosity of 0.60 to which 0.3 wt% of kaolin having an average particle diameter of 0.9 μm was added, dried at 170 ° C. for 3 hours, and then supplied to an extruder. For melt extrusion.

On the other hand, polypropylene containing 1.0 wt% of anatase type titanium and 0.05 wt% of tetra-n-butylphosphonium 3,5-dicarboxybenzenesulfonate as a polyolefin raw material for the core layer (melting temperature Tm: 152 ° C., melt flow rate: 5 g / 10
The mixture was dried at 100 ° C. for 1 hour using pellets of ethylene copolymerized polypropylene (ethylene copolymerized amount: 3 mol%), fed to another extruder, and melt-extruded at the same temperature as polyethylene terephthalate at 280 ° C. .

The respective molten polymers are merged in a die to form a three-layer multilayer structure of polyethylene terephthalate / polypropylene / polyethylene terephthalate, then discharged from a die, and then an electrostatic charge is applied to a cooling drum maintained at 20 ° C. While being wound, it was cooled and solidified to obtain a three-layer unstretched multilayer film. The unstretched multilayer film was heated to 80 ° C. by contacting it with a heating roll, stretched 3.6 times in the longitudinal direction, and immediately cooled to 20 ° C. Subsequently, 9 was performed in the transverse direction using a tenter-type transverse stretching apparatus.
After stretching 3.9 times at 0 ° C, heat-treat at 120 ° C,
After cooling to room temperature, it was wound up.

In the biaxially stretched multilayer film thus obtained, the average thickness of the surface polyethylene terephthalate layer was 1.5 μm, and the average thickness of the core polypropylene layer was 5 μm.
m.

In producing the film, the thickness of all layers was measured with a β-ray thickness meter as a radiation transmission type thickness meter for measuring all layers. Also, as an optical interference type film thickness meter for measuring the thickness of the surface layer, a film of a halogen lamp is irradiated to the film through an optical fiber, and the reflected light is reflected by a coaxial optical fiber into a spectrophotometer (specifically, Otsuka Electronics ( Co., Ltd., a spectrophotometer MCPD-2000), and an optical interference waveform generated at that time is subjected to Fourier transform and decomposed into a thickness component.
A configuration was used in which the thickness at which the maximum peak was taken was taken as the thickness of the film surface, which was provided on both sides of the film 5 as shown in the figure, and the surface layers on both sides were measured. This measurement was performed online, and the thickness of each layer was adjusted based on these data. In addition, the width direction position of the film of the β-ray head and the width direction position of the optical fiber film are synchronized,
It was made possible to measure at the same time and at the same place in the width direction.

The thickness data in the width direction of the film measured by each thickness gauge was input to a controller composed of a computer, and the core layer thickness at each measurement position was calculated by the following equation.
Core layer thickness measurement value = total layer thickness measurement value-surface layer thickness measurement value
In the core layer, the output for the operation end composed of the heat bolt was calculated, and by outputting each operation amount to the operation end, a film having good thickness unevenness of each layer in the width direction could be obtained.

Table 1 shows the values of the process gains Aij of the process models used for the control simulation means of the control means in the controller. In the table, the own layer is the layer to be controlled.

[0041]

[Table 1]

From Table 1, the above-described linear multilayer model in this example is given by the following equation.

[0043]

[Equation 3] ΔH1 (n) = 0.5 ΔOH1 (n) +0.3 ΔOR (n) +
0.1 △ OH2 (n) △ R (n) = 0.2 △ OH1 (n) + 0.4 △ OR (n) + 0.2 △ OH
2 (n) △ H2 (n) = 0.1 △ OH1 (n) + 0.3 △ OR (n) + 0.5 △ O
H2 (n) Table 2 shows the values of the interference rates Bk, n.

[0044]

[Table 2]

From Table 2, the linear interference model in this example is as follows.

[0046]

H1 (n) = 0.2 (H1 (n-2) + 0.5 △ H1 (n-1)
+ △ H1 (n) + 0.5 △ H1 (n + 1) + 0.2 △ H1 (n + 2) R (n) = 0.3 △ R (n-2) + 0.6 △ R (n-1 ) + △ R (n)
+ 0.6 △ R (n + 1) + 0.3 △ R (n + 2) H2 (n) = 0.2 △ H2 (n-2) + 0.5 △ H2 (n-1) + △ H2
(n) +0.5@H2 (n + 1) +0.2@H2 (n + 2)

At this time, a process gain for each operation end of each layer,
The interference rate should be input, but the measurement and input are complicated, and if the same layer, ignoring the variation did not greatly affect the controllability. did.

The unevenness of the thickness unevenness in each layer thickness profile was 20% before the adjustment, and the time required to fall within 5% after the adjustment was 8 hours in the conventional method. On the other hand, as a result of forming a film by the present adjustment method, the unevenness in thickness unevenness could be converged within 5% in one hour.

[0049]

As described above, according to the present invention, the conventional multi-point control means spends a great deal of time adjusting the thickness profile of each layer of the multilayer film. The thickness can be controlled, and the yield can be dramatically improved.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of a basic configuration of an embodiment of the present invention.

FIG. 2 is a block diagram of a basic configuration of thickness control.

FIG. 3 is a flowchart of a control unit according to the embodiment.

FIG. 4 is a flowchart of a process model according to the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Die 2 Surface thickness actuator 3 Core layer thickness actuator 4 Cooling drum 5 Film 6 Drawing device 7 Total thickness gauge 8 Surface thickness gauge 9 Winding device 10 Controller

 ──────────────────────────────────────────────────続 き Continued on front page (72) Inventor Tsukasa Nishimura 3-37-19 Koyama, Sagamihara-shi, Kanagawa F-term in Sagamihara Research Center, Teijin Limited 4F207 AA04 AA24 AG01 AG03 AP11 AQ01 AQ03 AR12 KA01 KA17 KB22 KL65 KL76 KL84 KM06 KM15

Claims (8)

[Claims] 1. A plurality of operation ends for adjusting a thickness of a die provided for a predetermined width disposed over the entire width of a die for molding a film from a molten resin, and corresponding to at least each of the operation ends in the film width direction. A thickness gauge for measuring the thickness of the film at each measurement point, and control means for controlling each operation terminal from each thickness measurement value measured at a predetermined cycle and each preset target value, and manufactured. In the film thickness control device configured to control the thickness of the film to the target value, the control means performs a predetermined control operation set based on a process model consisting of a mathematical model from the thickness measurement value and the target value. A control simulation means for simulating a control result, wherein the control end is controlled by a control output obtained by the control simulation means. Characteristic film thickness control device. The control simulation means includes convergence determination means for determining whether or not the calculated control result has converged, and the first control result obtained based on the input thickness measurement value and target value is determined. If the convergence determining means does not determine convergence, the first control result is used as the next thickness measurement value to obtain a second control result, and the procedure for determining by the convergence determining means determines that the control result is converged by the convergence determining means. The film thickness control device according to claim 1, wherein the calculation is repeatedly performed until the determination is made, and a control output when the convergence is determined is output to the operation terminal. 3. The film thickness control device according to claim 1, wherein said process model is a static linear first-order model. 4. The film thickness control device according to claim 3, wherein the process model is a linear interference model obtained by approximating mutual interference between a plurality of neighboring operation ends by a linear expression using a coefficient of interference as a coefficient. 5. The film is a multilayer film having two or more layers, and the process model is a linear multilayer model obtained by approximating the influence of the operation end of another layer on the layer by a linear equation using a process gain as a coefficient. The film thickness control device according to claim 4. 6. A control unit comprising state determination means for determining whether or not each of the thickness measurement values is in a steady state, wherein each of the thickness measurement values input in a predetermined cycle is determined to be in a steady state. At this time, the control output to the operation terminal is kept as it is, and when it is determined that the control output is not in the steady state, the control output obtained by the control simulation means from each of the thickness measurement values is output to the operation terminal. The film thickness control device according to claim 1. 7. The film thickness control device according to claim 6, wherein the convergence determination means also functions as a state determination means. 8. The film thickness control device according to claim 7, wherein the convergence determination means determines convergence when the measured thickness value or the control result of the simulation is within a set threshold value.