JP2662197B2 - Thickness control device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a thickness control device for controlling the thickness of a molten polymer material, for example, a film or sheet.

[0002]

2. Description of the Related Art In extrusion molding for producing a film or sheet, it is necessary to produce a product in which the thickness of the film or sheet is made constant. Therefore, conventionally, as described in Japanese Patent Publication No. 6-75908, for example, a plurality of heaters are arranged in the longitudinal direction of the die slot, and the observer estimates the state of the detected die lip temperature using a control algorithm. Then, the temperature of each heater is controlled based on the state estimation.

[0003]

However, in the above conventional example, since the temperature of the heater is controlled by estimating the state by the control algorithm, it is difficult to compensate for the temperature interference of each heater which actually occurs. There was a problem. Further, in the above conventional example, state estimation is performed using an observer. However, it is difficult to select an observer gain matrix L, and there is a problem that an error is likely to occur when performing estimation. Furthermore, in the above-mentioned conventional example, regardless of the instantaneous temperature change in the middle,
Since the thickness of the sheet is adjusted by reducing the estimation error with the passage of time, a gap may be generated between the actual temperature and the estimated temperature, and the thickness distribution may not be constant.

The present invention has been made in view of the above problems, and has a constant thickness distribution of a polymer material (hereinafter, referred to as “resin”) melted using an ARX (autoregressive exogenous) model. It is an object to provide a thickness control device.

[0005]

According to a first aspect of the present invention, a heating means comprising a plurality of heaters is provided in a width direction of a slit formed by a T-die lip, and the temperature of the heater is controlled. Accordingly, in the thickness control device for controlling the thickness of the molten polymer material (hereinafter, referred to as “resin”), the thickness control device is disposed near the discharge port of the T-die lip corresponding to each of the heaters. Temperature detecting means comprising a plurality of temperature sensors for detecting temperature, and calculating means comprising a state variable calculating unit for calculating a state variable of the heater based on an autoregressive exogenous model, using the detected temperature as a variable, There is provided a thickness control device including: a multivariable temperature control unit including a controller unit that controls the temperature of the heater according to the calculated state variable.

According to a second aspect of the present invention, a heat insulating means such as ceramic is provided between the heaters.

[0007]

According to the present invention, a heater state variable is calculated using an ARX model, the temperature of the heater is controlled in accordance with the state variable, and the thickness distribution of the resin discharged from the discharge port of the T-die lip is made constant. I do. According to the present invention, the heat interference between the heaters is reduced by insulating the heaters with ceramic, and the amount of heat from the heaters is supplied to the T-die.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A thickness control device according to the present invention will be described with reference to FIGS. FIG. 1 is a configuration diagram showing a conceptual configuration of extrusion molding, and FIG. 2 is a perspective view showing a configuration of the T-die shown in FIG. 1 and 2, a T-die 12 is attached to the extruder 11, and the T-die 1
2 is provided with die lips 14 and 15 forming a slit-shaped discharge port 13. A plurality of heaters 16 are arranged at equal intervals in the width direction above the die lip 14 and below the die lip 15, and a heat insulating material 17 made of ceramic or the like is arranged between the heaters 16. A temperature sensor 1 corresponding to each heater 16 is provided above the die lip 14 and near the tip of the discharge port 13.
8 for detecting the temperature at the tip of the discharge port 13. The temperature sensor 18 is, for example, a thermocouple.

Therefore, in this embodiment, the heat interference between the heaters 16 can be physically reduced by the heat insulating material 17 provided between the heaters 16, and the heat interference from the temperature sensors 18 is small. The temperature at the tip of the discharge port 13 can be detected. In the present embodiment, the heater 16 and the temperature sensor 18 are configured by five each. However, the number of the components is determined by the film or sheet to be created (hereinafter, “sheet etc.”).
That. ) It can be arbitrarily changed according to the width of 19.

The resin melted by the extruder 11 is sent to a temperature-controlled T-die 12, flows in the width direction of the T-die lip in a manifold 20, and after the resin pressure in the manifold 20 becomes sufficient, a discharge port is formed. The sheet 13 is extruded from the sheet 13 into a sheet 19 having a predetermined thickness, and is wound by a winder (not shown). FIG. 3 is a block diagram showing one embodiment of the configuration of the thickness control device according to the present invention. In the figure, T die 1
2 is heated by a heater 16, and each temperature sensor 18 detects the T-die temperature y at the tip of the discharge port 13 of the T-die 12.
(n) (where n is an integer) is detected and output to the state variable calculator 21 and the controller 22.

The state variable calculator 21 calculates the detected T
The state variable X (n) of the heater 16 is calculated based on the ARX model using the die temperature y (n) as a multivariable. That is, the state variable calculator 21 calculates the T-die temperature y (n) from each heater 16 and the heater input (input variable) u (n + 1) described later.
Are sequentially taken in and stored as past values, and an ARX model is constructed using the time series data as in the following equation. Here, in order to simplify the explanation, the estimation is performed by a T-die temperature system of a two-input two-output system.

First, the relationship between the input and output of the temperature system is represented by an ARX model: y1 (n) = a1y1 (n-1) + a2y1 (n-2) + a3y1 (n-3) + a4u1 (n-1) + a5u2 (n-1) (1) where y1 (n): detected temperature at a certain event (time) n y1 (n-1): detected temperature at a certain event n-1 y1 (n-2): certain Detected temperature at event n-2 y1 (n-3): detected temperature at certain event n-3 u1 (n-1): one heater input at certain event n-1 u2 (n-1): certain The other heater input at event n-1 a1, a2, a3, a4, a5: parameter y2 (n) = b1y2 (n-1) + b2y2 (n-2) + b3y2 (n-3) + b4u1 (n-1) + B5u2 (n-1) (2) where y2 (n): detected temperature at a certain event n y2 (n-1): detected temperature at a certain event n-1 y2 (n-2): certain event n -2 detected temperature y2 (n-3): detected temperature at a certain event n-3 b1, b2, b3, b4, b5: parameter Here, y1 (n), y1 (n-1), y1 (n-2), y1 (n-
3) and y2 (n), y2 (n-1), y2 (n-2) and y2 (n-3) are T
Since it is the temperature at the die, it can be directly detected by the temperature sensor.

Next, assuming the state variables as follows: X1 (n) = y1 (n) (3) X2 (n) = y1 (n-1) (4) X3 (n) = y1 ( n−2) (5) X4 (n) = y2 (n) (6) X5 (n) = y2 (n−1) (7) X6 (n) = y2 (n−2) (8) From the equation (1) and the equations (3) to (5), X2 (n + 1) = y1 (n) = X1 (n) X3 (n + 1) = y1 (n-1) = X2 (n )

[0014]

(Equation 1)

Equations (2) and (6) to
From (8), X5 (n + 1) = y2 (n) = X4 (n) X6 (n + 1) = y2 (n-1) = X5 (n)

[0016]

(Equation 2)

Is derived. Then, the above equation (9),
The following is a summary of (10).

[0018]

(Equation 3)

## EQU1 ## Therefore, it is possible to observe the temperature state of the process (T die) 12 from the above equation (11). Here, based on the above principle, the case of the embodiment shown in FIG. 3 will be considered. Since the present embodiment is a T-die temperature system having a 5-input / 5-output system, the relationship between the input and output of the temperature system in the state variable calculator 21 is as follows.

[0020]

(Equation 4)

Where a1 to a5, b1 to b5, c1 to c5,
d1 to d5, e1 to e5: each parameter u1 (n) to u5 (n): each heater input at a certain event n y1 (n) to y5 (n): each detected temperature at a certain event n X1 (n) ~ X15 (n): State variable here,

[0022]

(Equation 5)

[0023]

(Equation 6)

Then, the state variable X (n) in the above equation (12)
The state equation of (+1) and the target temperature y (n) is as follows: X (n + 1) = ΦX (n) + 状態 u (n) y (n) = CX (n) (13) According to this equation (13), it is possible to observe the temperature state X (n + 1) in the process (T-die). In FIG. 3, the output (state variable) of the state variable calculation unit 21 is X
(n), which expresses the temporal relationship with the heater input u (n + 1) from the controller unit 22 for convenience, and (n + 1) indicates the current state of the event. (N) shows the state immediately before that. Therefore, when expressing the current state variable of the state variable calculation unit 21, it is X (n + 1).

Accordingly, as shown in FIG. 3, the state variable calculator 21 calculates the state variable X (n) using the temperature detected by the temperature sensor 18 as a multivariable, and outputs the calculated state variable X (n) to the controller 23. That is, in the present invention, for a T-die temperature system of m input m output system (m is an arbitrary integer), A
When calculating the state variable of the heater based on the RX model, the state variable can be estimated as in the above embodiment.

The controller section 22 has a function of a subtractor, and subtracts a preset temperature set value yr from the detected T-die temperature y (n), and the subtraction result (state variable) e (n) e (n) = yr−y (n) = Δy (n) (14)

The controller 22 calculates a heater input u (n + 1) of the heater 16 in accordance with the input state variable X (n), and controls the temperature of each heater 16 based on the heater input. That is, the controller unit 22 stores past state variables and heater inputs to be input, and based on the past state variables and heater inputs, formulates a state equation of a state variable and a change amount of the heater input as follows: ΔX (n) = X (n + 1) −X (n) (15) Δu (n) = u (n + 1) −u (n) (16) Here, from Equations (13) to (16), e (n + 1) = yr-y (n + 1) = yr-CX (n + 1) = yr-C [.DELTA.X (n) + X (n) ] = Yr−CΔX (n) −CX (n) = e (n) −CΔX (n) (17) ΔX (n + 1) = X (n + 2) −X (n + 1) = ΦX (n + 1) + Γu (n + 1) −ΦX (n) −Γu (n) = ΦΔX (n) + ΓΔu (n) (18) When rearranging equations (17) and (18),

[0028]

(Equation 7)

When rearranging the above equation (14),

[0030]

(Equation 8)

## EQU1 ## Here, the evaluation function J = ΣΔy (n) ^T QΔy (n) + Δu (n) ^T RΔu (n) (21) where Q: positive or semi-positive (Q ≧ 0) R: positive (R> 0) In order to find the optimum control input Δu (n) that minimizes the following equation, it is necessary to solve the following Riccati equation.

P = Φ ^T PΦ + C ^T QC−Φ ^T PΓ (R + Γ ^T
PΓ) ^-1 Γ ^T PΦ From this positive definite symmetric solution P, the optimal control input Δu (n) is obtained as follows.

[0033]

(Equation 9)

Here, since e (n) finally converges to "0", the control guarantees that the temperature of each portion of the T-die will eventually reach the set value. Therefore, the controller unit 22 calculates the current heater input u (n + 1) u (n + 1) = Δu (based on the above equation (22) and the stored past value u (n) of the previous heater input. n) + u (n) (23) and outputs the heater input u (n + 1) to each heater 16 and the state variable calculator 21. Note that the state variable calculation unit 21 and the controller unit 22 can be configured by a CPU.

Next, the operation of the thickness control device shown in FIG. 3 will be described with reference to the flowchart of FIG. First, in the thickness control device, each parameter of the ARX model is set, and a control sampling time of the temperature near the discharge port of the T-die 12 is set (step 101).

Next, it is determined whether or not the above parameters are reset (step 102). This is because, for example, when extruding a material having a different material, the temperature detection accuracy becomes higher by resetting the parameters according to the material. Here, if the above parameters are to be reset, the process returns to step 101 and reset, and if the parameters are not reset, each temperature sensor 1
8, the temperature in the vicinity of the discharge port of the T-die 12 is detected.

Each time the control sampling time reaches the control sampling time (step 103) , the state variable calculating section 21 detects the temperature (state variable) y (n) detected by each temperature sensor 18 and the heater input output from the controller section 22. Read and store u (n + 1). Then, the state variable calculation unit 21
Each time series data and wherein the read (12), to calculate the state variables X (n) on the basis of (13), stores the state variables X (n), and outputs to the controller 22 (Step 10 4 ).

The controller unit 22 reads and stores the state variable y (n) and the state variable X (n) every time the control sampling time is reached. Then, the controller unit 22 compares the read time-series data of y (n) with the equation (1).
The subtraction result e (n) is calculated based on 4), and the state variable X (n) is calculated.
, An input variable (heater input) u (n + 1) is calculated based on the change amount e (n) and the equations (21) to (23), and the input variable u (n + 1) is stored ( In step 10 5 ), the temperature is output to the heater 16 and the state variable calculator 21 to control the temperature of each heater 16 (step 10 6 ).

Then, the thickness control device returns to step 103 and repeatedly controls the temperature of the heater 16 until, for example, an end instruction is given by the operator. Therefore, in the present embodiment, a heat insulating material is provided between the heaters to reduce the thermal interference between the heaters, the amount of heat from the heaters is given to the T-die side, and the detected temperature is used as a variable.
Since the heater input to each heater is controlled by multivariable temperature based on the RX model, the state variable e (n) converges to “0”,
Even if there is thermal interference between the heaters or external disturbance, the thickness distribution of the sheet or the like can be controlled precisely and at any time.

[0040]

As described above, according to the first aspect of the present invention,
In the thickness control device for controlling the thickness of the molten polymer material by providing a plurality of heating means in the longitudinal direction of the die slot, by controlling the temperature of the heating means,
A plurality of temperature detecting means disposed near the discharge port of the die slot corresponding to each of the heating means for detecting the temperature of the die slot; and A calculation unit for calculating the heating temperature of the heating unit based on the model; and a multivariable temperature control unit for controlling the temperature of the heating unit in accordance with the calculated heating temperature. The thickness distribution of the applied polymer material can be made constant.

According to the second aspect of the present invention, since the heat insulating material is provided between the respective heating means, the heat interference between the heaters is reduced by insulating the respective heaters, and the amount of heat from the heaters is supplied to the T-die side. it can.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing a conceptual configuration of extrusion molding according to the present invention.

FIG. 2 is a perspective view showing a configuration of a T-die shown in FIG.

FIG. 3 is a block diagram showing one embodiment of a configuration of a thickness control device according to the present invention.

4 is a flowchart for explaining the operation of the thickness control device shown in FIG.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Extruder 12 T die (process) 13 Discharge port 14, 15 Die lip 16 Heater 17 Insulation material 18 Temperature sensor 19 Sheet etc. 21 State variable calculation part 22 Controller part

Claims (2)

(57) [Claims] 1. A thickness control device for providing a plurality of heating means in a width direction of a slit formed by a die lip and controlling a temperature of the heating means to control a thickness of a molten polymer material. A plurality of temperature detecting means disposed near the discharge port of the die lip corresponding to the means, and detecting the temperature of the die lip; and using the detected temperature as a variable, the heating means based on an autoregressive exogenous model. A thickness control device comprising: calculation means for calculating a state variable; and multivariable temperature control means for controlling the temperature of the heating means in accordance with the calculated state variable. 2. The thickness control device according to claim 1, wherein the thickness control device includes heat insulation means provided between the heating means.